From 1efdabe2f4da19da482b3e6682e9e9d531889174 Mon Sep 17 00:00:00 2001
From: Jason Adams <jason.adams@automattic.com>
Date: Thu, 22 Oct 2026 20:26:19 -0600
Subject: [PATCH 1/7] feat: define one public benchmark release contract

---
 .../skills/wp-bench-execution-tests/SKILL.md  |    2 +-
 .github/workflows/ci.yml                      |   11 +-
 AGENTS.md                                     |    6 +
 README.md                                     |   33 +-
 benchmark/README.md                           |  151 ++
 benchmark/definition.json                     |  125 ++
 benchmark/releases.json                       |   11 +
 benchmark/releases/v1.json                    | 1723 +++++++++++++++++
 datasets/README.md                            |   16 +-
 .../execution/cli-content.json                |   14 +-
 .../execution/cli-maintenance.json            |   26 +-
 .../execution/cli-migration.json              |   10 +-
 .../execution/cli-options.json                |   18 +-
 .../execution/cli-reporting.json              |   14 +-
 .../execution/cli-taxonomy-meta.json          |   18 +-
 .../execution/cli-users.json                  |   22 +-
 .../execution/abilities-api.json              |   13 +-
 .../execution/admin.json                      |    3 +-
 .../execution/ai-client.json                  |   20 +-
 .../execution/auth.json                       |    3 +-
 .../execution/block-patterns-styles.json      |    3 +-
 .../execution/caching.json                    |    3 +-
 .../execution/comments-users.json             |    3 +-
 .../execution/connectors-api.json             |    7 +-
 .../execution/cron.json                       |    3 +-
 .../execution/database.json                   |    3 +-
 .../execution/dates-time.json                 |    3 +-
 .../execution/gb-block-api.json               |    3 +-
 .../execution/gb-block-bindings.json          |    3 +-
 .../execution/gb-block-editor.json            |    3 +-
 .../execution/gb-block-hooks.json             |    3 +-
 .../execution/gb-block-markup.json            |    3 +-
 .../execution/gb-font-library.json            |    3 +-
 .../execution/gb-interactivity-api.json       |    3 +-
 .../execution/gb-templates-navigation.json    |    3 +-
 .../execution/hooks.json                      |    3 +-
 .../execution/html-api.json                   |    3 +-
 .../execution/http-api.json                   |    3 +-
 .../execution/internationalization.json       |    3 +-
 .../execution/mail-url.json                   |    3 +-
 .../execution/media.json                      |    3 +-
 .../execution/performance.json                |    3 +-
 .../execution/plugin-lifecycle.json           |    3 +-
 .../execution/post-meta.json                  |    3 +-
 .../execution/post-types-taxonomy.json        |    3 +-
 .../execution/posts-lifecycle.json            |    3 +-
 .../execution/queries.json                    |    3 +-
 .../execution/rest-api.json                   |    3 +-
 .../execution/rewrite-permalinks.json         |    3 +-
 .../execution/roles-caps.json                 |    3 +-
 .../execution/scripts-styles.json             |    3 +-
 .../execution/security.json                   |    3 +-
 .../execution/settings-options.json           |    3 +-
 .../execution/shortcodes.json                 |    3 +-
 .../execution/theme-templates.json            |    3 +-
 python/tests/test_artifacts.py                |    4 +-
 python/tests/test_benchmark_releases.py       |  347 ++++
 python/tests/test_config_behavior.py          |    6 +-
 python/tests/test_continue_on_error.py        |    4 +-
 python/tests/test_dataset_filtering.py        |    6 +-
 python/tests/test_dataset_metadata.py         |   14 +-
 python/tests/test_execution_dataset.py        |    2 +-
 python/tests/test_execution_isolation.py      |    6 +-
 python/tests/test_execution_prompt.py         |    2 +-
 python/tests/test_exploit_audit.py            |    4 +-
 python/tests/test_reference_solution_mode.py  |    4 +-
 python/tests/test_result_schema.py            |   12 +-
 python/tests/test_result_streaming.py         |    4 +-
 python/tests/test_scoring.py                  |    2 +-
 python/tests/test_selection.py                |    6 +-
 .../tests/test_shell_runtime_integration.py   |    6 +-
 python/tests/test_skill_variants.py           |    4 +-
 python/tests/test_usage_capture.py            |    4 +-
 python/tests/test_wp_cli_shell.py             |   42 +-
 python/wp_bench/benchmark.py                  |  298 +++
 python/wp_bench/benchmark_cli.py              |  185 ++
 python/wp_bench/cli.py                        |    2 +
 python/wp_bench/config.py                     |   22 +-
 python/wp_bench/core.py                       |   23 +
 python/wp_bench/records.py                    |    5 +-
 wp-bench.example.yaml                         |   13 +-
 81 files changed, 3134 insertions(+), 241 deletions(-)
 create mode 100644 benchmark/README.md
 create mode 100644 benchmark/definition.json
 create mode 100644 benchmark/releases.json
 create mode 100644 benchmark/releases/v1.json
 rename datasets/suites/{wp-cli-v1 => wp-cli}/execution/cli-content.json (95%)
 rename datasets/suites/{wp-cli-v1 => wp-cli}/execution/cli-maintenance.json (96%)
 rename datasets/suites/{wp-cli-v1 => wp-cli}/execution/cli-migration.json (97%)
 rename datasets/suites/{wp-cli-v1 => wp-cli}/execution/cli-options.json (95%)
 rename datasets/suites/{wp-cli-v1 => wp-cli}/execution/cli-reporting.json (96%)
 rename datasets/suites/{wp-cli-v1 => wp-cli}/execution/cli-taxonomy-meta.json (95%)
 rename datasets/suites/{wp-cli-v1 => wp-cli}/execution/cli-users.json (96%)
 rename datasets/suites/{wp-core-v1 => wp-core}/execution/abilities-api.json (97%)
 rename datasets/suites/{wp-core-v1 => wp-core}/execution/admin.json (99%)
 rename datasets/suites/{wp-core-v1 => wp-core}/execution/ai-client.json (93%)
 rename datasets/suites/{wp-core-v1 => wp-core}/execution/auth.json (99%)
 rename datasets/suites/{wp-core-v1 => wp-core}/execution/block-patterns-styles.json (99%)
 rename datasets/suites/{wp-core-v1 => wp-core}/execution/caching.json (99%)
 rename datasets/suites/{wp-core-v1 => wp-core}/execution/comments-users.json (99%)
 rename datasets/suites/{wp-core-v1 => wp-core}/execution/connectors-api.json (98%)
 rename datasets/suites/{wp-core-v1 => wp-core}/execution/cron.json (99%)
 rename datasets/suites/{wp-core-v1 => wp-core}/execution/database.json (99%)
 rename datasets/suites/{wp-core-v1 => wp-core}/execution/dates-time.json (99%)
 rename datasets/suites/{wp-core-v1 => wp-core}/execution/gb-block-api.json (99%)
 rename datasets/suites/{wp-core-v1 => wp-core}/execution/gb-block-bindings.json (99%)
 rename datasets/suites/{wp-core-v1 => wp-core}/execution/gb-block-editor.json (99%)
 rename datasets/suites/{wp-core-v1 => wp-core}/execution/gb-block-hooks.json (99%)
 rename datasets/suites/{wp-core-v1 => wp-core}/execution/gb-block-markup.json (99%)
 rename datasets/suites/{wp-core-v1 => wp-core}/execution/gb-font-library.json (99%)
 rename datasets/suites/{wp-core-v1 => wp-core}/execution/gb-interactivity-api.json (99%)
 rename datasets/suites/{wp-core-v1 => wp-core}/execution/gb-templates-navigation.json (99%)
 rename datasets/suites/{wp-core-v1 => wp-core}/execution/hooks.json (99%)
 rename datasets/suites/{wp-core-v1 => wp-core}/execution/html-api.json (99%)
 rename datasets/suites/{wp-core-v1 => wp-core}/execution/http-api.json (99%)
 rename datasets/suites/{wp-core-v1 => wp-core}/execution/internationalization.json (99%)
 rename datasets/suites/{wp-core-v1 => wp-core}/execution/mail-url.json (99%)
 rename datasets/suites/{wp-core-v1 => wp-core}/execution/media.json (99%)
 rename datasets/suites/{wp-core-v1 => wp-core}/execution/performance.json (99%)
 rename datasets/suites/{wp-core-v1 => wp-core}/execution/plugin-lifecycle.json (99%)
 rename datasets/suites/{wp-core-v1 => wp-core}/execution/post-meta.json (99%)
 rename datasets/suites/{wp-core-v1 => wp-core}/execution/post-types-taxonomy.json (99%)
 rename datasets/suites/{wp-core-v1 => wp-core}/execution/posts-lifecycle.json (99%)
 rename datasets/suites/{wp-core-v1 => wp-core}/execution/queries.json (99%)
 rename datasets/suites/{wp-core-v1 => wp-core}/execution/rest-api.json (99%)
 rename datasets/suites/{wp-core-v1 => wp-core}/execution/rewrite-permalinks.json (99%)
 rename datasets/suites/{wp-core-v1 => wp-core}/execution/roles-caps.json (99%)
 rename datasets/suites/{wp-core-v1 => wp-core}/execution/scripts-styles.json (99%)
 rename datasets/suites/{wp-core-v1 => wp-core}/execution/security.json (99%)
 rename datasets/suites/{wp-core-v1 => wp-core}/execution/settings-options.json (99%)
 rename datasets/suites/{wp-core-v1 => wp-core}/execution/shortcodes.json (99%)
 rename datasets/suites/{wp-core-v1 => wp-core}/execution/theme-templates.json (99%)
 create mode 100644 python/tests/test_benchmark_releases.py
 create mode 100644 python/wp_bench/benchmark.py
 create mode 100644 python/wp_bench/benchmark_cli.py

diff --git a/.agents/skills/wp-bench-execution-tests/SKILL.md b/.agents/skills/wp-bench-execution-tests/SKILL.md
index 35d9581..fc5544b 100644
--- a/.agents/skills/wp-bench-execution-tests/SKILL.md
+++ b/.agents/skills/wp-bench-execution-tests/SKILL.md
@@ -151,7 +151,7 @@ The default `run.execution_isolation: reset_per_test` gives every candidate a fr
 
 ## WP-CLI Workflow Tests
 
-Use `wp-cli-v1` for realistic operational tasks that combine discovery, selection,
+Use `wp-cli` for realistic operational tasks that combine discovery, selection,
 changes, reports, and preservation. Prioritize model capability over testing tiny
 WordPress units. This suite targets single-site WordPress 7.1 with SQLite;
 avoid unsupported MySQL operations, network installs, or multisite instead of
diff --git a/.github/workflows/ci.yml b/.github/workflows/ci.yml
index f3124cf..f815c77 100644
--- a/.github/workflows/ci.yml
+++ b/.github/workflows/ci.yml
@@ -18,6 +18,8 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       - uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
 
       - uses: actions/setup-python@v5
         with:
@@ -33,6 +35,11 @@ jobs:
       - name: Type check (mypy)
         run: mypy python
 
+      - name: Benchmark release integrity
+        env:
+          BASE_REF: ${{ github.event.pull_request.base.sha || github.event.before }}
+        run: wp-bench benchmark check --base-ref "$BASE_REF"
+
       - name: Unit tests (pytest)
         run: pytest python -q
 
@@ -97,12 +104,12 @@ jobs:
 
       - name: WP-CLI workflow reference solutions
         run: >-
-          wp-bench run --config wp-bench.example.yaml --suite wp-cli-v1
+          wp-bench run --config wp-bench.example.yaml --suite wp-cli
           --check-reference-solution --execution-concurrency 4
 
       - name: WP-CLI workflow exploit audit
         run: >-
-          wp-bench run --config wp-bench.example.yaml --suite wp-cli-v1
+          wp-bench run --config wp-bench.example.yaml --suite wp-cli
           --check-exploits --execution-concurrency 4
 
       - name: SQLite database reference solutions
diff --git a/AGENTS.md b/AGENTS.md
index 66cf111..2b3250b 100644
--- a/AGENTS.md
+++ b/AGENTS.md
@@ -36,3 +36,9 @@ This guide is the fastest path to contributing to WP-Bench. Keep changes minimal
 - Follow Conventional Commits (`feat:`, `refactor:`, etc.) as seen in git history. Scope prefixes are optional.
 - PRs should state intent, key commands run (`ruff`, `mypy`, `pytest`, runtime smoke test), and config changes. Link related issues. Include logs or screenshots for runtime/verifier output when relevant.
 - Keep changes small and focused; update README/AGENTS.md when workflows or flags change. Avoid touching `legacy/` unless explicitly required.
+
+## Benchmark Releases
+- Public versions (`v1`, `v2`, …) identify the entire evaluation; `wp-core` and `wp-cli` are stable component IDs, never independent public versions.
+- Read `benchmark/README.md` before changing the release protocol, category mapping, result provenance, or website export contract.
+- Published `benchmark/releases/v*.json` files are immutable. Batch evaluation changes during development, then freeze the next release. Adding model results and changing presentation do not create benchmark versions.
+- Run `wp-bench benchmark check` and `pytest python/tests/test_benchmark_releases.py -q` after changes to versioning. Dataset/runtime/harness changes intentionally produce unreleased results until a new release is frozen.
diff --git a/README.md b/README.md
index 1a0f32e..b164877 100644
--- a/README.md
+++ b/README.md
@@ -2,6 +2,19 @@
 
 The official WordPress AI benchmark. Evaluate how well language models understand WordPress development—from core APIs and coding standards to plugin architecture and security best practices.
 
+## Public benchmark version
+
+One public version identifies the complete evaluation: **WP-Bench v1**. The
+`wp-core` and `wp-cli` datasets are stable components of that release. Changes to
+any dataset or score-affecting rules require the next benchmark version; adding
+model results updates the same version. Release manifests, run provenance and
+validated website exports are documented in [benchmark/README.md](benchmark/README.md).
+
+```sh
+wp-bench benchmark show --version v1
+wp-bench benchmark check
+```
+
 ## Overview
 
 WP-Bench measures AI model capabilities on **execution**: code generation tasks graded by static checks and runtime assertions in a real WordPress environment.
@@ -148,7 +161,7 @@ Copy `wp-bench.example.yaml` and customize:
 ```yaml
 dataset:
   source: local              # 'local' or 'huggingface'
-  name: wp-core-v1           # suite name
+  name: wp-core           # suite name
 
 models:
   - name: gpt-4o
@@ -160,7 +173,7 @@ grader:
   setup_timeout_seconds: 600 # hard cap for environment setup
 
 run:
-  suite: wp-core-v1
+  suite: wp-core
   limit: 10                  # limit tests (null = all); seeded stratified selection
   seed: 1337                 # selection seed (same seed = same subset)
   test_ids: []               # optional explicit test IDs to run
@@ -260,12 +273,12 @@ Test suites live in `datasets/suites/<suite-name>/`:
 
 - `execution/` — Code generation tasks with assertions (one JSON file per category)
 
-The default suite `wp-core-v1` contains 350 tasks covering WordPress core APIs,
+The default suite `wp-core` contains 350 tasks covering WordPress core APIs,
 hooks, database operations, and security patterns.
 
 ### WP-CLI workflows
 
-`wp-cli-v1` adds 27 operational tasks across content, taxonomy and metadata,
+`wp-cli` adds 27 operational tasks across content, taxonomy and metadata,
 users and capabilities, options and transients, reporting, maintenance, and migration.
 Tasks combine selection, changes, and preservation: examples include retiring
 an author while reassigning content, consolidating categories, repairing a
@@ -286,10 +299,10 @@ its research basis and source links in metadata. Prompts describe the desired
 WordPress outcome, and equivalent native command workflows are accepted.
 
 ```bash
-wp-bench run --config wp-bench.example.yaml --suite wp-cli-v1 --execution-concurrency 4
+wp-bench run --config wp-bench.example.yaml --suite wp-cli --execution-concurrency 4
 # Maintainer validation without model calls:
-wp-bench run --config wp-bench.example.yaml --suite wp-cli-v1 --check-reference-solution
-wp-bench run --config wp-bench.example.yaml --suite wp-cli-v1 --check-exploits
+wp-bench run --config wp-bench.example.yaml --suite wp-cli --check-reference-solution
+wp-bench run --config wp-bench.example.yaml --suite wp-cli --check-exploits
 ```
 
 The model returns a **plain command or Bash script string**, optionally inside
@@ -324,7 +337,7 @@ and repeat runs under `grader.raw.command`. Each output stream is capped at
 output is still recorded. Private container limits also apply, and timeouts
 or output overflow score zero.
 
-The WP-CLI suite is scored separately from `wp-core-v1`; existing PHP task
+The WP-CLI suite is scored separately from `wp-core`; existing PHP task
 contracts and scoring remain unchanged. Compare scores only within the same
 suite and execution profile.
 
@@ -378,11 +391,11 @@ tasks retain specialized assertions for their distinct contracts.
 dataset:
   source: huggingface
   name: WordPress/wp-bench-v1
-  suite: wp-core-v1       # filter a multi-suite export; defaults to run.suite
+  suite: wp-core       # filter a multi-suite export; defaults to run.suite
 ```
 
 An explicit `dataset.suite` selects the rows and their result-suite label;
-`--suite` overrides both. `--suite wp-cli-v1` selects CLI rows without changing
+`--suite` overrides both. `--suite wp-cli` selects CLI rows without changing
 the Hugging Face repository name. The new suite is available locally immediately;
 Hub users need an export that includes it.
 
diff --git a/benchmark/README.md b/benchmark/README.md
new file mode 100644
index 0000000..2597810
--- /dev/null
+++ b/benchmark/README.md
@@ -0,0 +1,151 @@
+# Public benchmark releases
+
+**WP-Bench v1** identifies one complete, frozen evaluation. `wp-core` and `wp-cli`
+are stable component IDs. They have no independent public versions. The Python
+package, result JSON schema, and scoring formula have internal version numbers;
+those are reproducibility details, not additional leaderboard versions.
+
+`definition.json` is the working evaluation recipe. `releases.json` indexes the
+append-only `releases/vN.json` manifests. A release freezes every dataset file,
+complete test membership, runtime and relevant harness source, WordPress target,
+scoring rules, broad category mapping, approved runtime images, generation
+settings, and required repeats. The manifest fingerprint covers the entire
+recipe. Changing any dataset changes the overall evaluation fingerprint.
+
+| Change | Release handling |
+| --- | --- |
+| Add/remove/edit a dataset task, including a corrected assertion | Next benchmark version |
+| Change grading, task weighting, prompting, or the evaluation protocol | Next benchmark version |
+| Change runtime behavior or the approved runtime images | Next benchmark version |
+| Evaluate another model on the same recipe | Updated results under the same version |
+| Re-run the same models under that recipe | Another results snapshot under the same version |
+| Edit documentation, website layout, or result serialization | Same benchmark version |
+
+Source fingerprinting is deliberately conservative. Even a dataset formatting or
+editorial-only edit changes its frozen file hash. Batch source changes during
+development rather than publishing a version for every commit. A changed checkout
+can run normally, but records `status: unreleased` and no public version. A new
+release is an explicit maintainer action; it is never inferred from directory
+names, a date, or a model list.
+
+## Running a release
+
+Inspect the complete protocol before spending on providers:
+
+```sh
+wp-bench benchmark show --version v1
+wp-bench benchmark check
+```
+
+An explicit assertion in the run config refuses mismatching sources, loaded
+local/Hub tasks, model settings, or runtime settings before model generation:
+
+```yaml
+benchmark:
+  version: v1
+
+dataset:
+  source: local
+  name: wp-core
+
+model:
+  name: YOUR_MODEL_ID
+  temperature: 1
+  max_tokens: 16384
+
+grader:
+  kind: docker
+  # Use one of runtime_image_ids from the frozen manifest, not an unverified tag.
+  image: sha256:6dcab391357137b2ce52d99ef15f342499b6b9f3287e5b96946036628a7dfb1e
+
+run:
+  suite: wp-core
+  execution_isolation: reset_per_test
+```
+
+Run both components and the required independent repeats. A subset remains
+useful for diagnostics but cannot enter the overall leaderboard. With Hugging
+Face, pin `dataset.revision` to the published export commit and select the stable
+suite ID. Loaded task fingerprints must match the frozen release regardless of
+transport. The historical Hub repository name `WordPress/wp-bench-v1` is an
+address, not a separate dataset-version authority; publish the new stable suite
+IDs there before using that export.
+
+Every completed payload carries `metadata.benchmark`; each streamed record also
+carries the benchmark identity, task fingerprint and `run_id`. Multi-model
+payloads retain a separate identity and run ID in each `models` entry. These
+fields survive partial runs, but publication requires actual complete records.
+A `full_suite` flag alone never authorizes a score. Source identity is captured
+before generation, so an edit during a run cannot retroactively relabel it.
+
+## Publishing a new benchmark version
+
+1. Batch intended evaluation changes in `definition.json`, datasets, and code.
+2. Run the normal Python/dataset checks and applicable reference/exploit audits.
+3. Freeze the next version, then inspect the generated manifest:
+
+   ```sh
+   wp-bench benchmark freeze v2
+   wp-bench benchmark check --base-ref origin/trunk
+   ```
+
+4. Commit the new manifest and catalog in the release PR. Existing manifests
+   cannot be edited, removed, or reassigned. CI compares their bytes with the base
+   commit, even if someone recomputes the embedded hashes.
+5. After merging, tag the release commit `benchmark-v2` and retain its approved
+   runtime images. To reproduce an old version, check out its tag; selecting an
+   old version alone does not restore its source into a changed checkout.
+
+If a changed file does not affect evaluation behavior, review that deliberately.
+The default fingerprint policy favors refusing an accidental comparison over
+silently declaring two evaluations equivalent. No file change should be hidden
+by modifying an old manifest.
+
+## Website data contract
+
+Use the harness exporter, rather than recreating release numbers in the website:
+
+```sh
+wp-bench benchmark export --version v1 --output-dir public-data \
+  --input core-model-a-repeat-1.json --input core-model-a-repeat-2.json \
+  --input cli-model-a-repeat-1.json --input cli-model-a-repeat-2.json
+```
+
+Supply the complete set of models desired in the new snapshot. Single- and
+multi-model result JSONs are supported. The exporter checks the release identity
+on both envelopes and individual records, task fingerprints, exact suite
+membership, approved runtime image, model/grader settings, baseline policy,
+unique run IDs, and the required independent repeats. Reference solutions,
+skills variants, errored/partial runs, duplicate repeats and mixed versions fail
+before output is created. A legitimate failed model answer is still a scored
+answer and remains in the denominator.
+
+The output is a small, stable ingestion contract, not a website theme:
+
+```text
+public-data/
+  results.json                # mutable index, grouped by benchmark version
+  v1/<snapshot-id>.json        # immutable scored-answer snapshots
+  v2/<snapshot-id>.json
+```
+
+Snapshot schema `1.0` contains `benchmark.version`, `benchmark.fingerprint`, the
+release-owned broad category labels, model IDs, per-answer pass/usage data, run
+IDs, input hashes, and an update date. It excludes generated code and raw provider
+responses. Every answer has a stable component ID, test ID and broad category.
+Aggregate pass rates weight every task equally and average the same number of
+repeats per task. Use task-paired uncertainty estimates when comparing models.
+
+The website should expose **one benchmark-version selector** and an **updated
+date**. Snapshot IDs and internal schemas belong in downloads/details. A model
+addition creates a new immutable snapshot under v1; changing any dataset creates
+v2 and its own leaderboard. Preserve v1 scores and files. Never fill missing v2
+rows with v1 scores. Cross-version charts must disclose test changes and cannot
+establish model improvement by themselves.
+
+The first release includes the already-audited four core assertion corrections
+and native read-only `wp db columns` allowance behind the initial public scores.
+Renaming component IDs and removing per-file versions changes no prompts or
+behavioral checks beyond those recorded corrections. Historical files without
+the new identity must be explicitly audited/imported for the first release;
+the exporter never guesses that an old `*-v1` suite name proves compatibility.
diff --git a/benchmark/definition.json b/benchmark/definition.json
new file mode 100644
index 0000000..3c17323
--- /dev/null
+++ b/benchmark/definition.json
@@ -0,0 +1,125 @@
+{
+  "schema_version": "1.0",
+  "wordpress_version": "7.1",
+  "scoring_version": "3.0",
+  "suites": [
+    "wp-core",
+    "wp-cli"
+  ],
+  "categories": [
+    {
+      "id": "core",
+      "label": "Core development",
+      "description": "Hooks, content, queries, settings, admin, internationalization, dates, plugins, and shortcodes."
+    },
+    {
+      "id": "blocks",
+      "label": "Block editor",
+      "description": "Blocks, patterns, bindings, interactivity, editor customization, and block markup."
+    },
+    {
+      "id": "themes-media",
+      "label": "Themes & media",
+      "description": "Theme templates, scripts, styles, and media handling."
+    },
+    {
+      "id": "apis",
+      "label": "APIs & integrations",
+      "description": "REST endpoints, HTTP, email, URLs, and scheduled tasks."
+    },
+    {
+      "id": "security",
+      "label": "Security & access",
+      "description": "Authentication, permissions, roles, capabilities, and safe handling of input and output."
+    },
+    {
+      "id": "modern",
+      "label": "New WordPress APIs",
+      "description": "Abilities, the AI Client, and Connectors APIs in the WordPress 7.1 runtime."
+    },
+    {
+      "id": "performance",
+      "label": "Data & performance",
+      "description": "Database access, caching, and performance tasks."
+    },
+    {
+      "id": "wp-cli",
+      "label": "WP-CLI",
+      "description": "Content, users, options, taxonomy, metadata, reporting, migration, and maintenance with WP-CLI."
+    }
+  ],
+  "category_mapping": {
+    "wp-core": {
+      "abilities-api": "modern",
+      "admin": "core",
+      "ai-client": "modern",
+      "auth": "security",
+      "block-patterns-styles": "blocks",
+      "caching": "performance",
+      "comments-users": "core",
+      "connectors-api": "modern",
+      "cron": "apis",
+      "database": "performance",
+      "dates-time": "core",
+      "gb-block-api": "blocks",
+      "gb-block-bindings": "blocks",
+      "gb-block-editor": "blocks",
+      "gb-block-hooks": "blocks",
+      "gb-block-markup": "blocks",
+      "gb-font-library": "blocks",
+      "gb-interactivity-api": "blocks",
+      "gb-templates-navigation": "blocks",
+      "hooks": "core",
+      "html-api": "core",
+      "http-api": "apis",
+      "internationalization": "core",
+      "mail-url": "apis",
+      "media": "themes-media",
+      "performance": "performance",
+      "plugin-lifecycle": "core",
+      "post-meta": "core",
+      "post-types-taxonomy": "core",
+      "posts-lifecycle": "core",
+      "queries": "core",
+      "rest-api": "apis",
+      "rewrite-permalinks": "core",
+      "roles-caps": "security",
+      "scripts-styles": "themes-media",
+      "security": "security",
+      "settings-options": "core",
+      "shortcodes": "core",
+      "theme-templates": "themes-media"
+    },
+    "wp-cli": {
+      "cli-content": "wp-cli",
+      "cli-maintenance": "wp-cli",
+      "cli-migration": "wp-cli",
+      "cli-options": "wp-cli",
+      "cli-reporting": "wp-cli",
+      "cli-taxonomy-meta": "wp-cli",
+      "cli-users": "wp-cli"
+    }
+  },
+  "aggregation": "strict pass rate; every task weighted equally; average independent repeats per task",
+  "protocol": {
+    "model": {
+      "temperature": 1.0,
+      "top_p": null,
+      "max_tokens": 16384
+    },
+    "grader": {
+      "kind": "docker",
+      "timeout_seconds": 90,
+      "container_cpus": 1.0,
+      "container_memory_mb": 512,
+      "container_filesystem_mb": 256,
+      "container_pids_limit": 64
+    },
+    "repeats": 2,
+    "reasoning_effort": "provider default",
+    "answer_policy": "first response; baseline without skills; transient provider errors retried; no scored-answer regeneration"
+  },
+  "runtime_image_ids": [
+    "sha256:6dcab391357137b2ce52d99ef15f342499b6b9f3287e5b96946036628a7dfb1e"
+  ]
+}
diff --git a/benchmark/releases.json b/benchmark/releases.json
new file mode 100644
index 0000000..d546de3
--- /dev/null
+++ b/benchmark/releases.json
@@ -0,0 +1,11 @@
+{
+  "schema_version": "1.0",
+  "releases": [
+    {
+      "version": "v1",
+      "path": "releases/v1.json",
+      "fingerprint": "83e81aa6ebc36f6516d0b38655f4818d556039a3507c03a029aef868c4171220"
+    }
+  ],
+  "current": "v1"
+}
diff --git a/benchmark/releases/v1.json b/benchmark/releases/v1.json
new file mode 100644
index 0000000..47283aa
--- /dev/null
+++ b/benchmark/releases/v1.json
@@ -0,0 +1,1723 @@
+{
+  "schema_version": "1.0",
+  "version": "v1",
+  "fingerprint": "83e81aa6ebc36f6516d0b38655f4818d556039a3507c03a029aef868c4171220",
+  "evaluation": {
+    "definition": {
+      "schema_version": "1.0",
+      "wordpress_version": "7.1",
+      "scoring_version": "3.0",
+      "suites": [
+        "wp-core",
+        "wp-cli"
+      ],
+      "categories": [
+        {
+          "id": "core",
+          "label": "Core development",
+          "description": "Hooks, content, queries, settings, admin, internationalization, dates, plugins, and shortcodes."
+        },
+        {
+          "id": "blocks",
+          "label": "Block editor",
+          "description": "Blocks, patterns, bindings, interactivity, editor customization, and block markup."
+        },
+        {
+          "id": "themes-media",
+          "label": "Themes & media",
+          "description": "Theme templates, scripts, styles, and media handling."
+        },
+        {
+          "id": "apis",
+          "label": "APIs & integrations",
+          "description": "REST endpoints, HTTP, email, URLs, and scheduled tasks."
+        },
+        {
+          "id": "security",
+          "label": "Security & access",
+          "description": "Authentication, permissions, roles, capabilities, and safe handling of input and output."
+        },
+        {
+          "id": "modern",
+          "label": "New WordPress APIs",
+          "description": "Abilities, the AI Client, and Connectors APIs in the WordPress 7.1 runtime."
+        },
+        {
+          "id": "performance",
+          "label": "Data & performance",
+          "description": "Database access, caching, and performance tasks."
+        },
+        {
+          "id": "wp-cli",
+          "label": "WP-CLI",
+          "description": "Content, users, options, taxonomy, metadata, reporting, migration, and maintenance with WP-CLI."
+        }
+      ],
+      "category_mapping": {
+        "wp-core": {
+          "abilities-api": "modern",
+          "admin": "core",
+          "ai-client": "modern",
+          "auth": "security",
+          "block-patterns-styles": "blocks",
+          "caching": "performance",
+          "comments-users": "core",
+          "connectors-api": "modern",
+          "cron": "apis",
+          "database": "performance",
+          "dates-time": "core",
+          "gb-block-api": "blocks",
+          "gb-block-bindings": "blocks",
+          "gb-block-editor": "blocks",
+          "gb-block-hooks": "blocks",
+          "gb-block-markup": "blocks",
+          "gb-font-library": "blocks",
+          "gb-interactivity-api": "blocks",
+          "gb-templates-navigation": "blocks",
+          "hooks": "core",
+          "html-api": "core",
+          "http-api": "apis",
+          "internationalization": "core",
+          "mail-url": "apis",
+          "media": "themes-media",
+          "performance": "performance",
+          "plugin-lifecycle": "core",
+          "post-meta": "core",
+          "post-types-taxonomy": "core",
+          "posts-lifecycle": "core",
+          "queries": "core",
+          "rest-api": "apis",
+          "rewrite-permalinks": "core",
+          "roles-caps": "security",
+          "scripts-styles": "themes-media",
+          "security": "security",
+          "settings-options": "core",
+          "shortcodes": "core",
+          "theme-templates": "themes-media"
+        },
+        "wp-cli": {
+          "cli-content": "wp-cli",
+          "cli-maintenance": "wp-cli",
+          "cli-migration": "wp-cli",
+          "cli-options": "wp-cli",
+          "cli-reporting": "wp-cli",
+          "cli-taxonomy-meta": "wp-cli",
+          "cli-users": "wp-cli"
+        }
+      },
+      "aggregation": "strict pass rate; every task weighted equally; average independent repeats per task",
+      "protocol": {
+        "model": {
+          "temperature": 1.0,
+          "top_p": null,
+          "max_tokens": 16384
+        },
+        "grader": {
+          "kind": "docker",
+          "timeout_seconds": 90,
+          "container_cpus": 1.0,
+          "container_memory_mb": 512,
+          "container_filesystem_mb": 256,
+          "container_pids_limit": 64
+        },
+        "repeats": 2,
+        "reasoning_effort": "provider default",
+        "answer_policy": "first response; baseline without skills; transient provider errors retried; no scored-answer regeneration"
+      },
+      "runtime_image_ids": [
+        "sha256:6dcab391357137b2ce52d99ef15f342499b6b9f3287e5b96946036628a7dfb1e"
+      ]
+    },
+    "files": {
+      "datasets/suites/wp-core/execution/abilities-api.json": "c449748609ef87f348eb2a57b1aaa1f9f52dae7d0e9e03e62a832cceccbe6708",
+      "datasets/suites/wp-core/execution/admin.json": "e494e305051950d5feb2e6a573685548ac105cfaf382cb84cef43151384ff759",
+      "datasets/suites/wp-core/execution/ai-client.json": "0f78a74c65ee3ab14d90f1841bb1ce6561b2dbf241dba5493399928b49593bf0",
+      "datasets/suites/wp-core/execution/auth.json": "fa876f3321aa30eac364c577ca60929851cde80e50dab5a505f909f64520539d",
+      "datasets/suites/wp-core/execution/block-patterns-styles.json": "a8ab2f6ab6f00d0e5abb8ad93bceafe604c6785d1fe81169a60417e28d74f5fc",
+      "datasets/suites/wp-core/execution/caching.json": "977ef8f4ce85c411b545acc42f1df774700668ca34e546fa66f0947e446f9777",
+      "datasets/suites/wp-core/execution/comments-users.json": "8f95820da861c60f49246bb49470edf12fa2adc3c9c2ebf4dbbee922aac3ced6",
+      "datasets/suites/wp-core/execution/connectors-api.json": "00244363a5ca965d2e8fa41e26e846f7d81f96636b2911a5c366f7d689019d83",
+      "datasets/suites/wp-core/execution/cron.json": "d80ee06220559f75e3c06cc960a540c37130bb6e128346e0e938a4b8470ff877",
+      "datasets/suites/wp-core/execution/database.json": "ccc32e70707ab93296d7759907a7eb824fbf60d6dc102692f66b797bd7c09c30",
+      "datasets/suites/wp-core/execution/dates-time.json": "fd4d2c7bebe9c746de66a50105422533bf0d21b084e967f73ce336395ca78f36",
+      "datasets/suites/wp-core/execution/gb-block-api.json": "1bd71b882a108baf44cba039782f8b993bcf168eede297e24b12f776c88f1626",
+      "datasets/suites/wp-core/execution/gb-block-bindings.json": "2cc4011318e56b543a0d9f65979b5fd709f13756a948bc7ea463b1168d65e57e",
+      "datasets/suites/wp-core/execution/gb-block-editor.json": "87d732f2693bb7beeeacff7bfc901bfc24e8acf827709eb4fc7fbabcd945ab73",
+      "datasets/suites/wp-core/execution/gb-block-hooks.json": "69daec04a64ccfad6911c9160c7b2f20ae6103d386fd42bf6d93c8f6edc8b43c",
+      "datasets/suites/wp-core/execution/gb-block-markup.json": "d49e2b15e2159b2f6e5e39784fc503dbaf6f1de909ae7f963c69dde77176dcb3",
+      "datasets/suites/wp-core/execution/gb-font-library.json": "9fbb0f37c11e7362bcd1670c10a7f49e24899525704d8641c327cecdf7bb624f",
+      "datasets/suites/wp-core/execution/gb-interactivity-api.json": "0c7e24c0c099ee545e8722a4d886a7b164983e53091c97071b5bfd32f273baa1",
+      "datasets/suites/wp-core/execution/gb-templates-navigation.json": "a044bc7506b1c5f7ae69f659435a995b3092aa6a2c1d806d015930bae3c6a965",
+      "datasets/suites/wp-core/execution/hooks.json": "f2e431af3a8c0b753f9e6cc93ab83e3d17264044f4ea44ec7b332578868e053a",
+      "datasets/suites/wp-core/execution/html-api.json": "4948bf9fbbbef6c7c486ddaef4668e22bcddb6aec778ad879c0afdd3991b7dfb",
+      "datasets/suites/wp-core/execution/http-api.json": "37a09fbb73837d456aa1942c6b3dd0b59a8c68daddca4dde14cc4d43e80308ca",
+      "datasets/suites/wp-core/execution/internationalization.json": "9f4cc060f6981504b3e9b7bacc53b7be9254ae9062635c1774c9b896db8c79ab",
+      "datasets/suites/wp-core/execution/mail-url.json": "2bae7cbbb3b794cac2d265ef61859e18aea06139b3efcd1c6238946a698514de",
+      "datasets/suites/wp-core/execution/media.json": "2427ab4935ba3a210c9724d07e4e54f796adc62ef016bf1039d408db6deca990",
+      "datasets/suites/wp-core/execution/performance.json": "b105c8e25adc825039d166df85fb9f433aeb28551e8624d441b830e67c3b51e5",
+      "datasets/suites/wp-core/execution/plugin-lifecycle.json": "98bd5bd0abd8b2944066b70c0f18d7a26ac7ae4bc457672360c3bd1b217d284a",
+      "datasets/suites/wp-core/execution/post-meta.json": "2b483afe5921f024bf753978ef013a9cf27591391ce3b5f81d929e4bdbd1cd44",
+      "datasets/suites/wp-core/execution/post-types-taxonomy.json": "ef207f34592bbf0780f575a815d44f92033f7a02b5861f43b40d2aaec9a0e98b",
+      "datasets/suites/wp-core/execution/posts-lifecycle.json": "03a5998f33d4b769b775c515ba56d3cd2fe9f861862eede8b43ad94ee7552561",
+      "datasets/suites/wp-core/execution/queries.json": "95f0a93e048d8d591c82aefaad0e7b12484be0c37eb3d7dc578c89adfb8f4d9b",
+      "datasets/suites/wp-core/execution/rest-api.json": "a948d0807fbdf3f45e3642531b1e732e4511d8d318362026dbd11b4ef3f440e2",
+      "datasets/suites/wp-core/execution/rewrite-permalinks.json": "7e41feda35a88954f107f47b0f94775d158e5b8ab9428af8bb17ee702549fa25",
+      "datasets/suites/wp-core/execution/roles-caps.json": "309323bfbe2001bc31e6e316d59e63024f562916dbad57d070d78af0b9739d18",
+      "datasets/suites/wp-core/execution/scripts-styles.json": "5f73409944d39813f4208de4b5c118ce6e6aa9580fb384cb0d8c15de7c6816f1",
+      "datasets/suites/wp-core/execution/security.json": "6ac099254eec381715b60c799ebc691e7ac02c0f9da496bc1788bc720972f66c",
+      "datasets/suites/wp-core/execution/settings-options.json": "9094bb1ff9e0570c878ff97bfe312dbd6dff19f06745f453bcb27a08056b7d2a",
+      "datasets/suites/wp-core/execution/shortcodes.json": "75601615067205696d0ac2feb59170ea73cc0d7b0aadffed6547254670f39e94",
+      "datasets/suites/wp-core/execution/theme-templates.json": "fc6b591a3b0b69b714645b04c50b29d2c533fc26c3a494eebe77b49269b7208e",
+      "datasets/suites/wp-cli/execution/cli-content.json": "e965ac79c813024a576d12a92f1be9d713bfc186b06e26518328a3df78308568",
+      "datasets/suites/wp-cli/execution/cli-maintenance.json": "873dfb05e7fe86df5d77b7a88408721fa223bf3aab97f3a9a4428319f1fabe1f",
+      "datasets/suites/wp-cli/execution/cli-migration.json": "25b178f3dde386acc62e990f2c28d5ac6d80c78099ad4a87f0bcd59d5284c1b7",
+      "datasets/suites/wp-cli/execution/cli-options.json": "ae573e7be2e23d12803cd43c0d98a7ac5bd7289386aa088e1780c6191092ae94",
+      "datasets/suites/wp-cli/execution/cli-reporting.json": "399cb90846e969ecea88b2a15d16189a7bc5c9e0201a5596e430cf3917f14af5",
+      "datasets/suites/wp-cli/execution/cli-taxonomy-meta.json": "5fe9afcd957fdd66448d85dcd3354ccd38f3dd1b9890c7f2d3b5a6f2cfdd44ee",
+      "datasets/suites/wp-cli/execution/cli-users.json": "e12f6331139a114606d59dbe9ed6b2e11857fef034140b8ad54f05b24424e583",
+      "runtime/Dockerfile": "b504512a5f989e38f1d44208f32b52581742b0ce5df34d72f3fabbf2fd98ffc7",
+      "runtime/docker-entrypoint.sh": "42a788d594951a5adb24f39da74c1f026d9d3ed92b5cc0cb26773290c484c317",
+      "runtime/record-loaded-files.php": "e65576fa8a4c32af088773793a0a67ee4fc1c11e166703f8d7f279b1c311c497",
+      "runtime/sqlite-snapshot.php": "3f69e126ed71d03e3840b62c08b72e8ba2b04ea133fe2eb69fc5713aad04dc13",
+      "runtime/src/class-artifact-installer.php": "cbee15169ea378ccf86e88e72a666f66563289432e1487fe16a54d50bbebecfa",
+      "runtime/src/class-preservation.php": "e24b9ba72f45008a00b9eac0dafa853cae864cd8c2a669c1f29b07c53c6f4861",
+      "runtime/src/class-sandbox.php": "ec5c1d95d511447ea25acb170efde1e838571062b02a412fc204fd6f3a7e331d",
+      "runtime/src/class-shell-verifier.php": "8f001f4cda8ff687b616c26139f1c6de7be6d83674395f09e7141dda0e5df17c",
+      "runtime/src/class-static-analysis.php": "368c5d17ee94401449237a7d37fa4ed47213e2b0979d56b9afc1eb00170da887",
+      "runtime/src/class-verifier.php": "f9962fd3f71becbf7951fe9ae23370aa0fdb5e46bc1fbd0d41392bcd79834e72",
+      "runtime/verify-runtime.php": "4fc6aad058e75f35b87bf236b9abb96c230998e820fa8bd3f61768e43d80362e",
+      "runtime/wp-bench-runtime.php": "2048b872b1d44cf94a9e710b847ad2242d5410597c66d33f533bb5f29985161a",
+      "python/wp_bench/artifacts.py": "e2d542a485ae14f5ad2c71681925d76ae2a374da526fa8ffdffc6786c1ea4179",
+      "python/wp_bench/config.py": "6ebcc19a98d58353baf7c8380ed9576a8cfcba97bef1fbb7688b364550a15ff5",
+      "python/wp_bench/utils.py": "53902e5ae682d0b4bfe8e9b031fd988374f4dd08ab25a976e08b3bf70c8aa30e",
+      "python/wp_bench/core.py": "30e9e518d24c94b114180bdc8a134e6d35bf17a2c4b3e7a7c7aa64524937bf65",
+      "python/wp_bench/datasets.py": "3d90bbc1fc6b96d739f8d8363ab69f31a3002856b4d9e3c11b8d4e3bfef9cca9",
+      "python/wp_bench/environment.py": "aad79cb091ccce35ebfc8c8f4f5b7bfd2ac12539742401d59b864256f2b72630",
+      "python/wp_bench/models.py": "f3a788544ffe8d23c0f3eec57abade81cd5b44b8b9d9ec6a2e48a73878568ada",
+      "python/wp_bench/scoring.py": "3d2b4664fe38741235960824a26156bcbbe32b046b39ae921d89b04e5761e91e",
+      "python/wp_bench/selection.py": "f486d2d960c3d159ce61b71024ea9487d52eff4dd7be0ebe1a441baf005ba16a",
+      "python/wp_bench/skills.py": "5d99dd457b45a6ee836f8ab46fcbd0693f89cfef14aafff027bd23b8252e548d",
+      "python/wp_bench/exploits.py": "f21b46043669b9f1d7ed1c78258a329c19f46803799f09ed1e9dcd73e2e391be"
+    },
+    "components": {
+      "wp-core": {
+        "task_count": 350,
+        "tasks": {
+          "e-abilities-api-001": {
+            "sha256": "f0b7a9198a50e40f08cdc9d4579e2614b4e4fe693df10153f9144cf9678b19ba",
+            "category": "abilities-api"
+          },
+          "e-abilities-api-002": {
+            "sha256": "7c7000f6e3128de66686acd22ec4b0b8adb42fca61fe0048509421ae225178c5",
+            "category": "abilities-api"
+          },
+          "e-abilities-api-003": {
+            "sha256": "93a5235867aac483660d7c98e4994b4191e53f2728bd061fc62f620b92077b62",
+            "category": "abilities-api"
+          },
+          "e-abilities-api-004": {
+            "sha256": "57723bc14ef830e6f5ed54f1ec0b46bef81e29482c4d2972dc068334e8e651d5",
+            "category": "abilities-api"
+          },
+          "e-abilities-api-005": {
+            "sha256": "53f74d879b76745863690468106cc6478d2daba6fa884b02a7c6c6b2298086c7",
+            "category": "abilities-api"
+          },
+          "e-abilities-api-006": {
+            "sha256": "d79dea07707e4ffef59a3528f922a30645e80c0e5cc19f7272367a051a1e655c",
+            "category": "abilities-api"
+          },
+          "e-abilities-api-007": {
+            "sha256": "838b3d8f115776492e6d0c8c533d24bb2c03bf6f62f53b1ea9797d0eea071f48",
+            "category": "abilities-api"
+          },
+          "e-abilities-api-008": {
+            "sha256": "f463f52c66092c9ba3bca32bdd106c5fc63fde8accdc7c16904cd2cf68f260f9",
+            "category": "abilities-api"
+          },
+          "e-abilities-api-009": {
+            "sha256": "d55f97d676ea098877c434af8417116537d174ff59e46aa2a94d2c0bf2e5938f",
+            "category": "abilities-api"
+          },
+          "e-abilities-api-011": {
+            "sha256": "5305117e02aaddab1eac19a1d424b0661e129df77ac52f8d1318b687c7f98898",
+            "category": "abilities-api"
+          },
+          "e-abilities-api-012": {
+            "sha256": "383291a210449db4cf968df2cd2c08cbd4817a522ed4a8c78b1bb0712305e774",
+            "category": "abilities-api"
+          },
+          "e-abilities-api-013": {
+            "sha256": "95ffe433eeafa9525634725bbe996f5626f70dd1a6a56de26ce93eda07aaa6a2",
+            "category": "abilities-api"
+          },
+          "e-abilities-api-014": {
+            "sha256": "78152b41ccc721cf7548e4927bdc99464404ca9f866f4b7852113cc8f9dd6904",
+            "category": "abilities-api"
+          },
+          "e-admin-001": {
+            "sha256": "7203857ddc667f16fb49cb1e1dd1e0c6655e31c0be5953c679bd02a3f5e17dff",
+            "category": "admin"
+          },
+          "e-admin-002": {
+            "sha256": "994939f9924b848eb6e11434c5901d0920d5b2ea345e7d3b61a207a84cdc3851",
+            "category": "admin"
+          },
+          "e-admin-003": {
+            "sha256": "1c731cb639b260d88120b1a26355bba92d324d67db82bfda4875a46112109244",
+            "category": "admin"
+          },
+          "e-admin-004": {
+            "sha256": "af8cbed7c0da0a61e6e96b96be157cbff334a25d4663d6141518e9aedbf457de",
+            "category": "admin"
+          },
+          "e-admin-005": {
+            "sha256": "e4cce76e32367a47c4e70d90cf2999b48ef3fd0b73927a5266ea67b747e18e4f",
+            "category": "admin"
+          },
+          "e-admin-006": {
+            "sha256": "61603606f30834c7bf33281e23ad843d44847b7011da3ae85e8eddb6e33483e5",
+            "category": "admin"
+          },
+          "e-admin-007": {
+            "sha256": "cccbb0d5270de0f7e0393856a39aeba33f5a54fc5fad30d024a816428138f9d6",
+            "category": "admin"
+          },
+          "e-admin-008": {
+            "sha256": "f491ed15036fa07a3382eb0eb2434935ba05232cca2513bc80c53d175e6c37ac",
+            "category": "admin"
+          },
+          "e-admin-009": {
+            "sha256": "bd41372ffe3fb8fbdb3cecaae6038e5e7377e9bcb9d7a8635d5d1b920862caff",
+            "category": "admin"
+          },
+          "e-admin-010": {
+            "sha256": "8c9476ffc99f2aaf38a766a58cdc34653117a203c3112fd879443163c1533fb0",
+            "category": "admin"
+          },
+          "e-admin-011": {
+            "sha256": "d5b5f2dd67df9a9e2bb3dd3df92810a278eed357d2be483b0970fb5d5eb359df",
+            "category": "admin"
+          },
+          "e-admin-012": {
+            "sha256": "f3d4220faf96d22292538b5faee3c37fc8f0146dd24391d20e4a44b487538fb8",
+            "category": "admin"
+          },
+          "e-ai-client-001": {
+            "sha256": "88c89c7471e9bf1383de5b6da6dc4f991ec2737a462a67df305a68f4f5744396",
+            "category": "ai-client"
+          },
+          "e-ai-client-002": {
+            "sha256": "ac0bd966fd50e8b245d11b59b3ba5238bb7a5c5313ea22b8cf8b4d4d4f3091bc",
+            "category": "ai-client"
+          },
+          "e-ai-client-003": {
+            "sha256": "0ed0faa53756de8c653c4a47f4efdf9c219c3b19b167267524b1458a35b1b08a",
+            "category": "ai-client"
+          },
+          "e-ai-client-004": {
+            "sha256": "f07d6d691cb9004d6fc1ace903e593fe264e28a11ff56c21b55be8489487a26a",
+            "category": "ai-client"
+          },
+          "e-ai-client-005": {
+            "sha256": "3d2aa8c9dca48fb7fa2dc0ab27293b94fb8ae67f7b919c41e3c9e5d3ec40edd9",
+            "category": "ai-client"
+          },
+          "e-ai-client-006": {
+            "sha256": "f20296b043005e0f823b346a9d4ee080dbd520060628bd2718c1439e9ae669c7",
+            "category": "ai-client"
+          },
+          "e-ai-client-007": {
+            "sha256": "8cd6f23f2b7d68c723e2fc03cac0d61d61889b2e72b5972a7f51638e932d8deb",
+            "category": "ai-client"
+          },
+          "e-ai-client-010": {
+            "sha256": "afc2060fa491db3e90f98e84088fd6e689db960076edb1236d71af36159d3221",
+            "category": "ai-client"
+          },
+          "e-ai-client-011": {
+            "sha256": "25c73d85aca695b84b640ac4f84d69f41a95e4cea51568cca6687473960456d5",
+            "category": "ai-client"
+          },
+          "e-ai-client-015": {
+            "sha256": "6a48b0855bcee04dd5ff45652a1e43fc5180314e431028fdedbc2aa7a5259f7b",
+            "category": "ai-client"
+          },
+          "e-auth-001": {
+            "sha256": "7837ee77326929ed4bf7df3d8bed18d66b8dd282ff6fc1d8e079d3a7a078e375",
+            "category": "auth"
+          },
+          "e-auth-002": {
+            "sha256": "4c0951721844d88afb0888a945f51a5a907bac797c53a998b3d51dae221be698",
+            "category": "auth"
+          },
+          "e-auth-003": {
+            "sha256": "010ff53f3ccc8b50dc239438392618ec655147adc2b44f008ce1a3e3d12a81b6",
+            "category": "auth"
+          },
+          "e-auth-004": {
+            "sha256": "7bd468058467747b724d63242b4c1499ef1137862c6b2d2d7572f3c7d1c7debe",
+            "category": "auth"
+          },
+          "e-auth-005": {
+            "sha256": "86e0d5b68673e096569948d20f8d27eaaa7662d84e526e38c9ce88d0995a1997",
+            "category": "auth"
+          },
+          "e-auth-006": {
+            "sha256": "7ec0252336430b9479eb1d07fdbb63070d2353e6b3e5074fb1aedc2ba9203f57",
+            "category": "auth"
+          },
+          "e-auth-007": {
+            "sha256": "b8995c7530bb9d834a645383b95ad98a8770ec15e5d490968ddfd0bf3cd4cb99",
+            "category": "auth"
+          },
+          "e-auth-008": {
+            "sha256": "91362bfd1d22f006bad8e564a61d829feb8d4a24614ac5e237cb35b88531bbf2",
+            "category": "auth"
+          },
+          "e-auth-009": {
+            "sha256": "a3ad8a2ada30bdc6d9a2dac2c1b61ccf9ea3ccecddb99edc060645695f3835fe",
+            "category": "auth"
+          },
+          "e-block-patterns-styles-001": {
+            "sha256": "2eddc347c6538eba9279f9993b3478f056fed959e98fd64db9215d7bdeca5a05",
+            "category": "block-patterns-styles"
+          },
+          "e-block-patterns-styles-002": {
+            "sha256": "57b1b86b5e6d06fad04d22581a1ec553258cd111099f14ba3240e90e3943c5de",
+            "category": "block-patterns-styles"
+          },
+          "e-block-patterns-styles-003": {
+            "sha256": "20f7d5387aeaa5803b6ae68fa5c23b9de380b0a20021648a445993f812489ace",
+            "category": "block-patterns-styles"
+          },
+          "e-block-patterns-styles-004": {
+            "sha256": "1130fd349afce650967f7afc9c680a0bdec0930cb28cfe860e2b797bfeb18410",
+            "category": "block-patterns-styles"
+          },
+          "e-block-patterns-styles-005": {
+            "sha256": "629d28a49250a1317fcdd5554be90f021f0a7ff00b6d941d43adbbb447dcf32c",
+            "category": "block-patterns-styles"
+          },
+          "e-caching-001": {
+            "sha256": "49d71affc4ec9f5e6f8eb7c073d161f044baef314dd3f443bc9f3c2e822acf83",
+            "category": "caching"
+          },
+          "e-caching-002": {
+            "sha256": "a7b6db8f1a76080b7d0f38176fea00818775e863598c966fa162ff2adcbcf371",
+            "category": "caching"
+          },
+          "e-caching-003": {
+            "sha256": "c2cf29995f4dbff7e5c9c931fca1148c0734911399fa0ca1936668e2fba1226a",
+            "category": "caching"
+          },
+          "e-caching-004": {
+            "sha256": "eedc7d787133d75a4d8da992214ddc2378d13327213eba75b9760083cdca6596",
+            "category": "caching"
+          },
+          "e-caching-005": {
+            "sha256": "5d3d11a21ca6a094e6d31e4269cba3dc628863851d1fd1022363d1c73154c373",
+            "category": "caching"
+          },
+          "e-caching-006": {
+            "sha256": "2e6bebb63568258f46ae4545910b3e39ee13cedaae1e558b724484b461ee6c0b",
+            "category": "caching"
+          },
+          "e-caching-007": {
+            "sha256": "a41138cbb365bc9db191c6d265a870658d18c71b2cabe1b065cec26c9c8e0ad1",
+            "category": "caching"
+          },
+          "e-caching-008": {
+            "sha256": "651fe2824b0787c16393887daa1725fced185285dc94c8b7d7c5bddac804f50b",
+            "category": "caching"
+          },
+          "e-caching-009": {
+            "sha256": "d3faec335333f47daa23c54af87ec84066c359caa3d3f3f4a88d69fde6b53458",
+            "category": "caching"
+          },
+          "e-comments-users-001": {
+            "sha256": "219bf20b686beca2c214d37c529c527839862702b7f3609b21e2cbb3c60a30b8",
+            "category": "comments-users"
+          },
+          "e-comments-users-002": {
+            "sha256": "eb2e78afa2f91932f0d2579881a88155fdcd4310902dcccc2d1df86324492cf8",
+            "category": "comments-users"
+          },
+          "e-comments-users-003": {
+            "sha256": "636280a7a3a7f13aeca5121dceb60623431411647a56ecbe3757ba5e7264dfd0",
+            "category": "comments-users"
+          },
+          "e-comments-users-004": {
+            "sha256": "04edd076018b7593e14088eba8eb2b781cd193eddb3140cb2642c090fd59b7cb",
+            "category": "comments-users"
+          },
+          "e-comments-users-005": {
+            "sha256": "570ed94a9de1ff0d4d0b3b21e3d9d52069c17b8dc8878dab9cfc7ec0dff93f16",
+            "category": "comments-users"
+          },
+          "e-comments-users-006": {
+            "sha256": "b628458ae28d6d318956bea170cd40c0a152a09a8fcb82f34a3a7d8e0cbbcffe",
+            "category": "comments-users"
+          },
+          "e-comments-users-007": {
+            "sha256": "b258ac238c14b48efc113637c3270f0b1c3b0ffd7e871dd2704a97c6dfd3f848",
+            "category": "comments-users"
+          },
+          "e-connectors-api-001": {
+            "sha256": "88fd0d229ed06764856c25e4d801cb82a465f5632579448f5dfb156d764e39be",
+            "category": "connectors-api"
+          },
+          "e-connectors-api-002": {
+            "sha256": "c221ff980bb918b39317952da4531ec56c589a67daa005c79e022a3291cd08b0",
+            "category": "connectors-api"
+          },
+          "e-connectors-api-003": {
+            "sha256": "59451f5a1626677285561a62f6ad4c76ccd491312b7bb93e75d94d40418fb45e",
+            "category": "connectors-api"
+          },
+          "e-connectors-api-004": {
+            "sha256": "502098d053a8963a5d97f29216984fb60a2b688644593423a14f9569c9a99640",
+            "category": "connectors-api"
+          },
+          "e-connectors-api-006": {
+            "sha256": "13e4dc178119c96ca0adb363176c285a43080e04b247902b74f0883198936646",
+            "category": "connectors-api"
+          },
+          "e-connectors-api-008": {
+            "sha256": "cdf52c5948b7d069ff1ad49837923cc924ee45a82611711fb4d22c28e364399d",
+            "category": "connectors-api"
+          },
+          "e-connectors-api-009": {
+            "sha256": "03095a89a152296925172198d6e318be8ab0b2186145f11fdd85d929c4b567e3",
+            "category": "connectors-api"
+          },
+          "e-connectors-api-010": {
+            "sha256": "f45d3683d7bf4889fcaeb4251807c5a52e281640d67862b339ff429c7b4f9335",
+            "category": "connectors-api"
+          },
+          "e-connectors-api-011": {
+            "sha256": "4ccfccfc73e9868a47b67e8da7439b900ff6192f715a5d8b89ef22cb0fbbb733",
+            "category": "connectors-api"
+          },
+          "e-connectors-api-013": {
+            "sha256": "122adca07ee95b93c5631dd7432a0db15cf51824c7b9922b22ea46ff02fd9476",
+            "category": "connectors-api"
+          },
+          "e-cron-001": {
+            "sha256": "82a89b9a2bf017d4c6b2a48c756132d4729f7de927c6e274ec7f45b8474ed866",
+            "category": "cron"
+          },
+          "e-cron-002": {
+            "sha256": "8c2a06c38db536f5c4cf6d2c4e365524b52f2af95c0d93bcc6ea67673d275986",
+            "category": "cron"
+          },
+          "e-cron-003": {
+            "sha256": "621f52a29071e0a174f03d811ed599a7a33e26bb7b66c1d77fd5db5544ef9aac",
+            "category": "cron"
+          },
+          "e-cron-004": {
+            "sha256": "3895a945375beec935af78cf04e90ecf14ceaeb245e26c54649989513d16cfe0",
+            "category": "cron"
+          },
+          "e-cron-005": {
+            "sha256": "e4144556b3a6f2ef6b1cd763abcab119185949f67442919058a436ace15a9038",
+            "category": "cron"
+          },
+          "e-cron-006": {
+            "sha256": "ac12653ac263b523ed47838479b37b761ec265d99de8adf395454b24a789bbcb",
+            "category": "cron"
+          },
+          "e-cron-007": {
+            "sha256": "2bdbaa37576948c06cb3469c940ba4e6f4bbfdbcb68291d4333528d224772f0a",
+            "category": "cron"
+          },
+          "e-database-001": {
+            "sha256": "1929351d46953ab765200bc69bc23ef9228e3057e5c1b01f965eddeb32e0736b",
+            "category": "database"
+          },
+          "e-database-002": {
+            "sha256": "b3c3c0ac052266855abc0a2ab2de2c812700d1e66d532d087a159ac6e7d6a78d",
+            "category": "database"
+          },
+          "e-database-003": {
+            "sha256": "ce954e9a38701dc7380f7b236788662bf6f90606667953e28452d1eb5495d739",
+            "category": "database"
+          },
+          "e-database-004": {
+            "sha256": "909b3094a2dcd7ff3f3c5f3ffadfb52fd87635d0b393dd7517b7a55921b06461",
+            "category": "database"
+          },
+          "e-database-005": {
+            "sha256": "b6b9f91556141857c2b8ebfc216889a39f9fe255f66832a3039ada6cb63af45f",
+            "category": "database"
+          },
+          "e-database-006": {
+            "sha256": "958b6e1f580b162f28eca2101fd2eba1135daaaa95105f467c99d9c5c7c07f2f",
+            "category": "database"
+          },
+          "e-database-007": {
+            "sha256": "317bb3729838c0236b16b43be51d3d28d149ae954f1a4bb05505b97e937aea82",
+            "category": "database"
+          },
+          "e-database-008": {
+            "sha256": "a0dc921ae2547c1d6f47ec16ad33267613c1f0cf9745323e4711f18d620ba057",
+            "category": "database"
+          },
+          "e-dates-time-001": {
+            "sha256": "43999e8178476f3c57456297ea96312dd4a87f14b57f3ab50b056b780265e1a3",
+            "category": "dates-time"
+          },
+          "e-dates-time-002": {
+            "sha256": "87299679beb13389977e537916e02d60d1f00e95a236c5d8c8ef4f52996b5d57",
+            "category": "dates-time"
+          },
+          "e-dates-time-003": {
+            "sha256": "836783c1acc9dc94a368e6fd9c99d8b522f8d50bc3414016294e6f426b1bea96",
+            "category": "dates-time"
+          },
+          "e-dates-time-004": {
+            "sha256": "1d253a7717cda0bbcf4c23c5a866aa0dbdcbb0267a572955d3f6b3bc187f26fa",
+            "category": "dates-time"
+          },
+          "e-dates-time-005": {
+            "sha256": "3140e1e7fadae051333d3f221aaa3333cdbc01a4dcb099f05ede42efab2ee946",
+            "category": "dates-time"
+          },
+          "e-dates-time-006": {
+            "sha256": "438859ac8d91bae827f0f1d6a4ee1a66fbaa7b0c006ce9c50535dfbb13cff3eb",
+            "category": "dates-time"
+          },
+          "e-gb-block-api-001": {
+            "sha256": "f8cc073f8aa62de612315caa5e54be8e9ccd3ca28d863deacf5f1576a019a2b9",
+            "category": "gb-block-api"
+          },
+          "e-gb-block-api-002": {
+            "sha256": "9dc48929edd30170f825169d24eac96583a5a923be0d744fe73a34fab8333944",
+            "category": "gb-block-api"
+          },
+          "e-gb-block-api-003": {
+            "sha256": "3a4ef7a068da883da2788cdc3b592688377c7e392a8501844f0f959f1908e99a",
+            "category": "gb-block-api"
+          },
+          "e-gb-block-api-004": {
+            "sha256": "14f557553bee50df87a3dc6c1123607beb8ed4334320449890ef830dc3faed6c",
+            "category": "gb-block-api"
+          },
+          "e-gb-block-api-005": {
+            "sha256": "986ef4300e03f106a70971ad89da4ae676b6b662f61d597d22a24aa70ecf4dbd",
+            "category": "gb-block-api"
+          },
+          "e-gb-block-api-006": {
+            "sha256": "65f8fb0ec0c414cd8b308e0dc489d97083ac15f504d8a038bea6ffbe7069244f",
+            "category": "gb-block-api"
+          },
+          "e-gb-block-api-007": {
+            "sha256": "01e9c174ff82fddf84ff0833614315f12697b793438cc2ef087a60b36b11cc4c",
+            "category": "gb-block-api"
+          },
+          "e-gb-block-api-008": {
+            "sha256": "883c5405a16809d76ed38505bfa1fc585b6409d0aeaf07745ecf0f364486768c",
+            "category": "gb-block-api"
+          },
+          "e-gb-block-api-009": {
+            "sha256": "e2befec1a0315aed852dc30f7da1f7b3d04d1d883b1b7cd9a4b1dd299e4c6c62",
+            "category": "gb-block-api"
+          },
+          "e-gb-block-api-010": {
+            "sha256": "9393f872e6d44b78b80287bbaf2dac961a7bf75d993d02090e6c082900eafb13",
+            "category": "gb-block-api"
+          },
+          "e-gb-block-api-011": {
+            "sha256": "4e1e1140ccf77a6d7924ed0dcb17241db2c2065e7aa4c77236fa7a78673d4713",
+            "category": "gb-block-api"
+          },
+          "e-gb-block-api-012": {
+            "sha256": "a1e2a8e73e1ed4919c9b5249f5af61702cc995ca4b1184911d0f1fc1111e41b7",
+            "category": "gb-block-api"
+          },
+          "e-gb-block-api-013": {
+            "sha256": "09becd2a1b96c8f37e8abfca018ae495bb2e4ab2aa9387f8a87263488c9970d3",
+            "category": "gb-block-api"
+          },
+          "e-gb-block-api-014": {
+            "sha256": "c580ab75fa0f372728e1f7a76a6a8883d074c7730dcf05aa14b06ce7ce38cb71",
+            "category": "gb-block-api"
+          },
+          "e-gb-block-api-015": {
+            "sha256": "1b0c1fe194e8576ad75298c50e0b91f6e8d7ac706716b3fcff9e34753f64256f",
+            "category": "gb-block-api"
+          },
+          "e-gb-block-api-016": {
+            "sha256": "ff7ed858de6b8e34016db615c43e3176b65ba9f72825ab18fd173d56432aadb6",
+            "category": "gb-block-api"
+          },
+          "e-gb-block-bindings-001": {
+            "sha256": "8b17e7def95f2f262f93aee13979b9c155e456080c102ad90bf49be9f3f2fd81",
+            "category": "gb-block-bindings"
+          },
+          "e-gb-block-bindings-002": {
+            "sha256": "aacc2324a42de95830730f8f49867fc66ba72b6a803f38559aa95bf158e2f059",
+            "category": "gb-block-bindings"
+          },
+          "e-gb-block-bindings-003": {
+            "sha256": "670af9861734083a5d6790b433d4faf4991cf1c4ff0e91fc7e2123c18be86ef9",
+            "category": "gb-block-bindings"
+          },
+          "e-gb-block-bindings-004": {
+            "sha256": "50b172d1a9736c4b94b73678eea6be96ef117d2491cd492735ee26fa20534b13",
+            "category": "gb-block-bindings"
+          },
+          "e-gb-block-bindings-005": {
+            "sha256": "88ea6c95db288f14ee8f7f3b159ac9259025622e28e839c3c8b62bce0fdea8a7",
+            "category": "gb-block-bindings"
+          },
+          "e-gb-block-bindings-006": {
+            "sha256": "44dda7169244841de5c7112cf72fa9cfb56491cb1745813a58e52b91a2cd24d4",
+            "category": "gb-block-bindings"
+          },
+          "e-gb-block-editor-001": {
+            "sha256": "52fd7ef7abf4f9ca41aa64dafcdbe5f32f5d66323b37108e35aa7d32e4c44413",
+            "category": "gb-block-editor"
+          },
+          "e-gb-block-editor-002": {
+            "sha256": "ca0ca4d30110404a3d27bdd26f07680337c675522c9d1cf40c06d7d312611783",
+            "category": "gb-block-editor"
+          },
+          "e-gb-block-editor-003": {
+            "sha256": "f1e7c6a7a1f71e89bc3000cc02201194090a9f22019d11f4483490a5f73d2006",
+            "category": "gb-block-editor"
+          },
+          "e-gb-block-editor-004": {
+            "sha256": "23ce260c07bf0a9cfe1c4e42d3d04281bd6c8d0fd46c84800100c5d9173d4277",
+            "category": "gb-block-editor"
+          },
+          "e-gb-block-editor-005": {
+            "sha256": "620dfcd38648df06ed8b647c3bda9be5b0b0bbe0709de0a755b739a34ce4de76",
+            "category": "gb-block-editor"
+          },
+          "e-gb-block-editor-006": {
+            "sha256": "c2d1b57cc8c7ceba601daf9870d8b590a10ae86209d3bc84db16cf8a4776d8b3",
+            "category": "gb-block-editor"
+          },
+          "e-gb-block-hooks-001": {
+            "sha256": "43c1e0819ae9004562d58b9c5fce134fd9712b1b6da6ab215fcef64421346e13",
+            "category": "gb-block-hooks"
+          },
+          "e-gb-block-hooks-002": {
+            "sha256": "388caece059cb61f5f9a7404e68ab21ff4195bd1cd7032402bbd89b0ea9e0b6c",
+            "category": "gb-block-hooks"
+          },
+          "e-gb-block-hooks-003": {
+            "sha256": "d42f4ffb4cfc32f0901cbeac425e7c733ac7fb40355d337f026249dae0e4a3e1",
+            "category": "gb-block-hooks"
+          },
+          "e-gb-block-hooks-004": {
+            "sha256": "a16c2f05f81a804c4d53a6569c9d543600495afd53182d43a4e450db96a7ce9e",
+            "category": "gb-block-hooks"
+          },
+          "e-gb-block-markup-001": {
+            "sha256": "a6a087d6cad8728c4e431abde258fec264cab808123b0fa51610b9c77546ebaf",
+            "category": "gb-block-markup"
+          },
+          "e-gb-block-markup-002": {
+            "sha256": "a0a33425bee8e1bf3b305afdb56a85739d222d05a547a2c394db7b211d44e9f5",
+            "category": "gb-block-markup"
+          },
+          "e-gb-block-markup-003": {
+            "sha256": "2ec07b89781c30a113828aa453a5273472336d9be9ca7414215190837842f87a",
+            "category": "gb-block-markup"
+          },
+          "e-gb-block-markup-004": {
+            "sha256": "a773d8a648960ead43335b020c9d75032bfd0a099b0831ab11b8642c11216723",
+            "category": "gb-block-markup"
+          },
+          "e-gb-block-markup-005": {
+            "sha256": "e0b12fcb692a98e6faad5edfc14d01cc27467d44e19eaebcabe2177ac474f623",
+            "category": "gb-block-markup"
+          },
+          "e-gb-block-markup-006": {
+            "sha256": "0d0ec7e514e5e7bda7519122d3dde216c03cb066e727c208165d8d16049b5079",
+            "category": "gb-block-markup"
+          },
+          "e-gb-block-markup-007": {
+            "sha256": "71868a9e886c2ab762546f9c035dcedb0e7238a80f14c6381a3ff2282f161d5c",
+            "category": "gb-block-markup"
+          },
+          "e-gb-block-markup-008": {
+            "sha256": "0aa49fdbdaed84971da2e44d9c245db6804b638f3cb6988c99fb0de7cbd62dde",
+            "category": "gb-block-markup"
+          },
+          "e-gb-block-markup-009": {
+            "sha256": "3e5992b6e09887835b750897bfc5841927746996b49addc5d01d412e73b20892",
+            "category": "gb-block-markup"
+          },
+          "e-gb-block-markup-010": {
+            "sha256": "4898a12dcba55520d69e294a8c50e0e881f8707d8cdac588af6db025260db91a",
+            "category": "gb-block-markup"
+          },
+          "e-gb-block-markup-011": {
+            "sha256": "5456bb751eb09a10df2d26422d51bf22596abb6720cbda86f9216fec180be1c1",
+            "category": "gb-block-markup"
+          },
+          "e-gb-block-markup-012": {
+            "sha256": "e20e5f76d9b596af066a1bd530c9343215168b2bbaad9842f872ecf1a954343c",
+            "category": "gb-block-markup"
+          },
+          "e-gb-block-markup-013": {
+            "sha256": "e01bf1e499efc7b5cc258ac56b38e7998936080dbeb15ad457a9e6461260a1dd",
+            "category": "gb-block-markup"
+          },
+          "e-gb-block-markup-014": {
+            "sha256": "4359cb583990d903259cc4de0d1520563869d1cff5bad1ff78bc7d8dbcd8c204",
+            "category": "gb-block-markup"
+          },
+          "e-gb-font-library-001": {
+            "sha256": "9175bb47f6e4006883e8748165500b488e14a722736fdd1ec5d4c3763b810514",
+            "category": "gb-font-library"
+          },
+          "e-gb-font-library-002": {
+            "sha256": "2ec520094b91ead39b3caf5a91487997dfeb0fbe76c20918ffe88c90a76ebdec",
+            "category": "gb-font-library"
+          },
+          "e-gb-font-library-003": {
+            "sha256": "fb1c4dfe6f2f865f3f6e455cd37e86136bc3d7b9bd9c7b3608c1a4cc3a83f45b",
+            "category": "gb-font-library"
+          },
+          "e-gb-interactivity-api-001": {
+            "sha256": "ce22414ee3041ca7f7c9d1af6fbd9d09fe091c34ca9af912d92e6588792f727d",
+            "category": "gb-interactivity-api"
+          },
+          "e-gb-interactivity-api-002": {
+            "sha256": "8c82e5ba14644f0e892fc8c1766c0bc5ff3cc1fefb317cf0cef5fad23bfe6923",
+            "category": "gb-interactivity-api"
+          },
+          "e-gb-interactivity-api-003": {
+            "sha256": "8821e5c9c939ba4f284e9072560c1c49630439980c033aca6866837bd93cd1bc",
+            "category": "gb-interactivity-api"
+          },
+          "e-gb-interactivity-api-004": {
+            "sha256": "843dc6df92a7b393322a74c91a67b7a64d60a74d054d09050e204494ac3757b2",
+            "category": "gb-interactivity-api"
+          },
+          "e-gb-interactivity-api-005": {
+            "sha256": "045ffd36a945c144e8c0e2a97393c23dd570175b4f291b173b99514ae9ce011a",
+            "category": "gb-interactivity-api"
+          },
+          "e-gb-interactivity-api-006": {
+            "sha256": "85f7002f86c0f4817b37c97af69786c7f860903e1a857ceda5482c3f62f394a1",
+            "category": "gb-interactivity-api"
+          },
+          "e-gb-interactivity-api-007": {
+            "sha256": "96ac27aae295d6532afd0e8ca9cf0992b0ed28f4f70e0aa716328e475e039c09",
+            "category": "gb-interactivity-api"
+          },
+          "e-gb-interactivity-api-008": {
+            "sha256": "1b833832d8316f2addb03379b2813b132ed078a99cea38589ec9ad18deacd683",
+            "category": "gb-interactivity-api"
+          },
+          "e-gb-interactivity-api-009": {
+            "sha256": "9862f1b67050de54ca0d7dd63647740aa6b14fb6318d017b0cf4b7131d6fe1d7",
+            "category": "gb-interactivity-api"
+          },
+          "e-gb-interactivity-api-010": {
+            "sha256": "5c820b7c83da82e8e359610f2fec0230ff53d2565a695866509bd230f2a89231",
+            "category": "gb-interactivity-api"
+          },
+          "e-gb-interactivity-api-011": {
+            "sha256": "c4fa4334b2725facb89094579edebc5cd53b81c4874ad04a5bc3626ebc6cd162",
+            "category": "gb-interactivity-api"
+          },
+          "e-gb-templates-navigation-001": {
+            "sha256": "4e7224549b6f17f82fcece5205fe12cd7942a2551673dee88147a7acac8eb086",
+            "category": "gb-templates-navigation"
+          },
+          "e-gb-templates-navigation-002": {
+            "sha256": "4ee8e5f44e8d26cd69eb946def6afe7c0ebe2c19f1dcb56a7bc56f86f305e8b0",
+            "category": "gb-templates-navigation"
+          },
+          "e-gb-templates-navigation-003": {
+            "sha256": "dbc706ada6e54f7a5f8c4a6465a13b3c507b08b75090e49bc616b3d6476f80d3",
+            "category": "gb-templates-navigation"
+          },
+          "e-hooks-001": {
+            "sha256": "fdbf40ed5e90a1fd1f4c0b821d1ac51a04bfaef1b754cc8b3770ada3ebdc4788",
+            "category": "hooks"
+          },
+          "e-hooks-002": {
+            "sha256": "64bd91b63cfef0fc2c8abc19641d6220e4d8912911b903c5b2cf2d388a110832",
+            "category": "hooks"
+          },
+          "e-hooks-003": {
+            "sha256": "5acd616f402032b24cb50a743a0d8737f4c4ef84ecc7026cec3bb0728124f445",
+            "category": "hooks"
+          },
+          "e-hooks-004": {
+            "sha256": "705e6d00a48aae394fd11a555c69aa0a42a50275757bcdf3f8c52f3cf0043d70",
+            "category": "hooks"
+          },
+          "e-hooks-005": {
+            "sha256": "3d74ab83fad0fc386b2d8f4041532f79fc9393cb05726f9fe66855b653ed5b0e",
+            "category": "hooks"
+          },
+          "e-hooks-006": {
+            "sha256": "4023ac4c5912361768e09f92d3f1b916267943657b7975af765035e8103dbee8",
+            "category": "hooks"
+          },
+          "e-hooks-007": {
+            "sha256": "d35bc795ba08b31b8c10c08daf96ea4208cc0f58297b362054878a726cb1db9f",
+            "category": "hooks"
+          },
+          "e-hooks-008": {
+            "sha256": "7e3a695ee050f9e7033df7a123da977db3a006ebc23f3e4da445c4e69f96e324",
+            "category": "hooks"
+          },
+          "e-html-api-001": {
+            "sha256": "ae9a3dc35798c8d81cff6a5dc3971cc45627eed3180b2cd864d76307d9e0ca73",
+            "category": "html-api"
+          },
+          "e-html-api-002": {
+            "sha256": "03d2fabd0e5a8b0d94ebc190a147702f26dcc83722adb127c277d2d6493d875f",
+            "category": "html-api"
+          },
+          "e-html-api-003": {
+            "sha256": "c2634aeb795c1c773abfcb85e1be3b5fbda946778de7c493e9cec82c570ca8d5",
+            "category": "html-api"
+          },
+          "e-html-api-004": {
+            "sha256": "f3d13a19a7641ad242bdbacdff85a2cff2ab2403f8008af144dcdffef69ee5ef",
+            "category": "html-api"
+          },
+          "e-html-api-005": {
+            "sha256": "a6d57b8bf95098f9b91f31c86c49aba2771fabbbf5753ace5d1768a892c2012d",
+            "category": "html-api"
+          },
+          "e-html-api-006": {
+            "sha256": "e81b0d271f68e0a556003d6698e0e3b918aa9684d2308d08540084fafb30f53e",
+            "category": "html-api"
+          },
+          "e-html-api-007": {
+            "sha256": "6e967b5d727f905b26be73d80c65dc29ee6c8ab7e5bcbdbb6ca9dabfecbff8bb",
+            "category": "html-api"
+          },
+          "e-html-api-008": {
+            "sha256": "9234ae62d5a760450268c0a02e56c0de92b308717bf19103d2a09f84a204a17d",
+            "category": "html-api"
+          },
+          "e-html-api-009": {
+            "sha256": "a023ab121f752644970242994d11b85964b7d12eec1a4a861337b7d2551e537d",
+            "category": "html-api"
+          },
+          "e-html-api-010": {
+            "sha256": "50936d425764a304104d07d59bc4eeb3a6c82fd4ff7c82a66953dc1d6cbfc275",
+            "category": "html-api"
+          },
+          "e-http-api-001": {
+            "sha256": "afce26b1848b5fe254a3d2f914c3ffdd4bf0966d9036c76697f1a0f27e05119f",
+            "category": "http-api"
+          },
+          "e-http-api-002": {
+            "sha256": "14433b68d51b7d1787b707ec037204e92380ad93885e06948cab2c6e8e803150",
+            "category": "http-api"
+          },
+          "e-http-api-003": {
+            "sha256": "67a702839a3eaa90b0a9cfbb060860f114d83de2692ecb033fb540f07d6157ed",
+            "category": "http-api"
+          },
+          "e-http-api-004": {
+            "sha256": "dabd6ed0f97fa023cbe360f7564959dad8b4d03728eb0e283ec4b6ccdb1df8fb",
+            "category": "http-api"
+          },
+          "e-http-api-005": {
+            "sha256": "34a1e6ef947fd50f824822576dac133b090d14e8f2fcfb53085da378c76a7a8c",
+            "category": "http-api"
+          },
+          "e-http-api-006": {
+            "sha256": "ab668a3240c865b9b54a4e47c4e6c683eed2704634acf5eddc41cad3d29a76af",
+            "category": "http-api"
+          },
+          "e-http-api-007": {
+            "sha256": "e98ecc5a61dad8c7bd57edd469cd3b44f6817e08edf58b7b74d036f68f114fc9",
+            "category": "http-api"
+          },
+          "e-http-api-008": {
+            "sha256": "a49228cbca205c971a6ffe5ed7be99186546338dbe5adaa2973dd0f3963fda5b",
+            "category": "http-api"
+          },
+          "e-internationalization-001": {
+            "sha256": "43dd4f043e2d8f15b121aaa5f64f38b7897e3b5c16351b589026083504cbcede",
+            "category": "internationalization"
+          },
+          "e-internationalization-002": {
+            "sha256": "7390f00dbb1a67b43c4f8fa9d739f986ea1a548f7757c19aba5a333e8caa44ae",
+            "category": "internationalization"
+          },
+          "e-internationalization-003": {
+            "sha256": "3116acfe7e6ec848726842132db3ef914d5aa19f17cedadc9f784096ea9fc792",
+            "category": "internationalization"
+          },
+          "e-internationalization-004": {
+            "sha256": "b256532c7a2ab8b4ef2f467eed36bbe241353b9309fad348c65f6a2705ad17f7",
+            "category": "internationalization"
+          },
+          "e-internationalization-005": {
+            "sha256": "5e57ad18f4697a53082ff0a425ad9837053d9513d5ce70d88a9dbe1d3f39fca1",
+            "category": "internationalization"
+          },
+          "e-internationalization-006": {
+            "sha256": "96e73daef4f570f54cf972df42cd2c732302e5d81fd87359664a86decbeb2763",
+            "category": "internationalization"
+          },
+          "e-mail-url-001": {
+            "sha256": "6070e535c2423236048ba66440e83460b1e50d977de7686bea68d8a245846968",
+            "category": "mail-url"
+          },
+          "e-mail-url-002": {
+            "sha256": "a00c33f3f27577341cc7983f079bc20c6b63ab29715da9194ff4be3ab5a61c69",
+            "category": "mail-url"
+          },
+          "e-mail-url-003": {
+            "sha256": "80679c4904ddcd6276fbee74aa4785d66a6519681cdd8f6ded6a6d29f74d2393",
+            "category": "mail-url"
+          },
+          "e-mail-url-004": {
+            "sha256": "6a158d379a6bbfd102d690e02108ca36b7dee3ee65abd9642a6ef10c55e5617d",
+            "category": "mail-url"
+          },
+          "e-mail-url-005": {
+            "sha256": "3ae117eff7944c6a3d5c34c766b8f91f4960de3836c83edb3c4becec32ad4e78",
+            "category": "mail-url"
+          },
+          "e-mail-url-006": {
+            "sha256": "9d1ec712b75acc806813860191dc2d3242b56ff663f58ec78127553711090080",
+            "category": "mail-url"
+          },
+          "e-mail-url-007": {
+            "sha256": "2fbdce59645583ee09957d9d962b1aec8aab480a7ed2aca07a155428477c54cc",
+            "category": "mail-url"
+          },
+          "e-media-001": {
+            "sha256": "4fb806f46b50758eb1137dea81b97bfd77f7d10e0003522c015394bb17bcaf1f",
+            "category": "media"
+          },
+          "e-media-002": {
+            "sha256": "bd21ad7ea4b84f2397040a0966ad599c28a94044b92eb72725d70ba78d0a6a5f",
+            "category": "media"
+          },
+          "e-media-003": {
+            "sha256": "168bb3b888a4a96d740cb8279864998c79f997665f4e390bb317e3d528f5683c",
+            "category": "media"
+          },
+          "e-media-004": {
+            "sha256": "2ba990a143425af283b469b988312d9581ac9b2ec6b9b470a40a32ea94b139fa",
+            "category": "media"
+          },
+          "e-media-005": {
+            "sha256": "869db99c30f136709639ec9a5b5ed684f22e5aefc0d0206ff24959e5f023e141",
+            "category": "media"
+          },
+          "e-media-006": {
+            "sha256": "bf71a04216505f83cca58994b9eed57182bc38c6275c01962510cb0ef7ea458b",
+            "category": "media"
+          },
+          "e-media-007": {
+            "sha256": "aed5365ebb743da8929f505a4486757e04e30183112d4f39520e8cf21bd4cfb7",
+            "category": "media"
+          },
+          "e-media-008": {
+            "sha256": "cc3a680bffaae639ec90e13427b3e84798e317bd64d391d355ed1a605323298d",
+            "category": "media"
+          },
+          "e-media-009": {
+            "sha256": "ded6188407772a7bfe46459823a84dca614b6887a0bd1329b4a89d8361540f78",
+            "category": "media"
+          },
+          "e-performance-001": {
+            "sha256": "d1ce8cd064c32f4408dd0b692fcd6346b421178309bd4cbcb6ff7a50c9222b6f",
+            "category": "performance"
+          },
+          "e-performance-002": {
+            "sha256": "17db6298c2df5368025c9b3488cafbb9887a646d9d86c58845b8d61e535962d3",
+            "category": "performance"
+          },
+          "e-performance-003": {
+            "sha256": "baff3d17e02ecc59526a4ba2c5cf19f487ccdfb3a8af1a04005a161ca2664d69",
+            "category": "performance"
+          },
+          "e-performance-004": {
+            "sha256": "3a3d8cb7e1e30f54f74699a256c814c9a8bf2c6561a0a9401b4e038f2dee9a53",
+            "category": "performance"
+          },
+          "e-performance-005": {
+            "sha256": "239863a1f7659bd6a795adda79b678c369db0d2128520de3df2505f5945a2946",
+            "category": "performance"
+          },
+          "e-performance-006": {
+            "sha256": "54a336e213f64e3f5ff4967a1f60e2349dd1d450b31e8f8f2750ee745a2c4b75",
+            "category": "performance"
+          },
+          "e-performance-007": {
+            "sha256": "15365a1ba03131d1a5f8dcebc46e91794c82bced99c71377e88d82f87f187f9c",
+            "category": "performance"
+          },
+          "e-plugin-lifecycle-001": {
+            "sha256": "6fa5fb59a3765613f0fb901eb2e172ff0d4b55b398996cc3faa21704761646d1",
+            "category": "plugin-lifecycle"
+          },
+          "e-plugin-lifecycle-002": {
+            "sha256": "58680765787719aa96c45252635ca8858d8a8bf3d426555132e89b19f9503f77",
+            "category": "plugin-lifecycle"
+          },
+          "e-post-meta-001": {
+            "sha256": "3983010e8dfdc201b8326da32e7beedb0361f5be4cf5b01c7450c28a192d9986",
+            "category": "post-meta"
+          },
+          "e-post-meta-002": {
+            "sha256": "258f03374a98fddc518fbc2efa8f781abb6754ea774a561ac615c1babedcf2ca",
+            "category": "post-meta"
+          },
+          "e-post-meta-003": {
+            "sha256": "a8df5bef82fd40590704c2520ffbeaf76495f7d6c8d4fa5cdfec1427ecd0ceea",
+            "category": "post-meta"
+          },
+          "e-post-meta-004": {
+            "sha256": "7a81322e17df7cd22de11c825bf731dbd8e094e4b35cc2361e3b678bc684bcd8",
+            "category": "post-meta"
+          },
+          "e-post-meta-005": {
+            "sha256": "ae0a06a7bd81822f12bd290e4f4e267ead53831a5514e6aa528689060af5bcde",
+            "category": "post-meta"
+          },
+          "e-post-meta-006": {
+            "sha256": "9011ab8c64911e800539ab8a07105338ea8fac3d599623afcd90b9edd6fd7529",
+            "category": "post-meta"
+          },
+          "e-post-types-taxonomy-001": {
+            "sha256": "4d99b0c3bcc10bc1d1789c7f1b2fc3f6f7524d62f9d3fc2eb734cb88ff167dd7",
+            "category": "post-types-taxonomy"
+          },
+          "e-post-types-taxonomy-002": {
+            "sha256": "a1485fa4e681046a33129d36896165f784fa21a168b13d32d419f4987035ec7b",
+            "category": "post-types-taxonomy"
+          },
+          "e-post-types-taxonomy-003": {
+            "sha256": "dced4db39bce3c84e8fc167a342674fb4d611e756a463f7da4f72647a8794e94",
+            "category": "post-types-taxonomy"
+          },
+          "e-post-types-taxonomy-004": {
+            "sha256": "c769d66fa33a988abc0ee833c528d336f11744c09bf00da981175561193b8272",
+            "category": "post-types-taxonomy"
+          },
+          "e-post-types-taxonomy-005": {
+            "sha256": "ebaf3b98cf1c2ce4da008bbbe98cfe229980d34841a4bc80f9b45e0fa238e2e2",
+            "category": "post-types-taxonomy"
+          },
+          "e-post-types-taxonomy-006": {
+            "sha256": "4fcdc928ffff98b12f291ee5b353b65ef76363875ddd3b95f0b690189406d822",
+            "category": "post-types-taxonomy"
+          },
+          "e-post-types-taxonomy-007": {
+            "sha256": "73c2c17ac717f7eeaa2ef876d3136e04ebc0d25797a617810c52f384c15ec4da",
+            "category": "post-types-taxonomy"
+          },
+          "e-post-types-taxonomy-008": {
+            "sha256": "2d6a7f5333fb14dbdb5672fa7ad9cdc938cb388f8eceddea7464adae53b9d6b9",
+            "category": "post-types-taxonomy"
+          },
+          "e-post-types-taxonomy-009": {
+            "sha256": "c1a519195596b17d8e06775b2d68c75b3d97ebe298d6ff009d8afc66b0f533c4",
+            "category": "post-types-taxonomy"
+          },
+          "e-post-types-taxonomy-010": {
+            "sha256": "85900e1f4c9a6a8c16d1785ed8888a8528326f86556105d7f064d2b3ec098c98",
+            "category": "post-types-taxonomy"
+          },
+          "e-post-types-taxonomy-011": {
+            "sha256": "c025e824eb5bd38c42007769833ea4ee548c2eef6fbdfeb8efbaa22d73bb80e3",
+            "category": "post-types-taxonomy"
+          },
+          "e-post-types-taxonomy-012": {
+            "sha256": "376429fcdf62f8f4c4f7d3469fcd44908919e45116513075e4362aa0f0d7f195",
+            "category": "post-types-taxonomy"
+          },
+          "e-post-types-taxonomy-013": {
+            "sha256": "19b1c7683b5b9eb8e2a8823461f99211942dea4d7a67cde71f657545a6234388",
+            "category": "post-types-taxonomy"
+          },
+          "e-post-types-taxonomy-014": {
+            "sha256": "3526f93d417807469fefa062f86e495faa551e5ef1f798cb69508559d89870d9",
+            "category": "post-types-taxonomy"
+          },
+          "e-posts-lifecycle-001": {
+            "sha256": "1f0455301f64c69b5fd9a207309dde7c4b4bd4848316181a37df18a3bd4487bb",
+            "category": "posts-lifecycle"
+          },
+          "e-posts-lifecycle-002": {
+            "sha256": "944c08bd5e48a82ca0f269294b025dcacdcc7ceba7aef39f77c5d8ab01d46f3c",
+            "category": "posts-lifecycle"
+          },
+          "e-posts-lifecycle-003": {
+            "sha256": "220f446a4a020fb8812d6393623f6d3d7bf0b923fc6d0ff72d20a0a0fc5855e2",
+            "category": "posts-lifecycle"
+          },
+          "e-posts-lifecycle-004": {
+            "sha256": "a659a204417357319be42748c3a936e03d7600e7a800fd9d243f10d068abdb9a",
+            "category": "posts-lifecycle"
+          },
+          "e-posts-lifecycle-005": {
+            "sha256": "dcce8d2921a1ffc8ad7ced4f640462da9195450a0b6b29ee7b410571bcb757eb",
+            "category": "posts-lifecycle"
+          },
+          "e-posts-lifecycle-006": {
+            "sha256": "df9ece1797a791a8327be7dcd67ca6048039aafcd0ccee44446dd83e4561ed90",
+            "category": "posts-lifecycle"
+          },
+          "e-posts-lifecycle-007": {
+            "sha256": "2e4c89e23b522ac055f48f2df5db2c6f9af8be3620afbed1e01328cda851b3cb",
+            "category": "posts-lifecycle"
+          },
+          "e-posts-lifecycle-008": {
+            "sha256": "1c23c811d11997994a26ce31cd2e75f8531d8422ad4a35666bcb5983c9f1e789",
+            "category": "posts-lifecycle"
+          },
+          "e-posts-lifecycle-009": {
+            "sha256": "7dfcd742bede510bb6ee08dad9cb159fb0305c06cf0ff2aed340f6adfd909045",
+            "category": "posts-lifecycle"
+          },
+          "e-posts-lifecycle-010": {
+            "sha256": "20350eb1b1ca1aa1cfe3c0c6a70677d7d8b8903a6fade9a5133a98d8f9e7e078",
+            "category": "posts-lifecycle"
+          },
+          "e-posts-lifecycle-011": {
+            "sha256": "b955bb2a19947150de620f523f130e946852907f586946d98b51fc7826ec4bbc",
+            "category": "posts-lifecycle"
+          },
+          "e-posts-lifecycle-012": {
+            "sha256": "c7908eacff97783281c4d9b463e1639f5d73d206868fea72649414f2594a2e7a",
+            "category": "posts-lifecycle"
+          },
+          "e-queries-001": {
+            "sha256": "9fb5e4d6469ffe0b21b14b55c003db44e0ed85cf7b3d866f2a64d02e02c3f34e",
+            "category": "queries"
+          },
+          "e-queries-002": {
+            "sha256": "e07f4289c72f0d4e865565780d93df9a21623336ad3bd70d10a248e602bbbbae",
+            "category": "queries"
+          },
+          "e-queries-003": {
+            "sha256": "e4375af0b046bc708a3664640317504032570ad22a7ff443053886d6a72144ba",
+            "category": "queries"
+          },
+          "e-queries-004": {
+            "sha256": "05124f74f45817617418ace69b2a06e70ee7e7d81d5cb9ff176fbf6ad0c6352b",
+            "category": "queries"
+          },
+          "e-queries-005": {
+            "sha256": "260f491ac70491e05d5b2a9f48a3f3d8818669180a12ef320913a1e018a9207e",
+            "category": "queries"
+          },
+          "e-queries-006": {
+            "sha256": "f6355351c542979390d32d8997196925b61cb4cc8cfffa388c906f9df0aea2ba",
+            "category": "queries"
+          },
+          "e-queries-007": {
+            "sha256": "af79700b283644a6fb6797b7b7c9d6d0012b3db0b3f95afb51d026a723797cf5",
+            "category": "queries"
+          },
+          "e-queries-008": {
+            "sha256": "bc7b0cc2c250960d0df86a5fb32bb77be72055f1f42c6b3ebacb29720e60d239",
+            "category": "queries"
+          },
+          "e-queries-009": {
+            "sha256": "0ddf4d9667e223dd086eeda0e16e73bc1edf8a9300a850fe7121b86cfe72db19",
+            "category": "queries"
+          },
+          "e-queries-010": {
+            "sha256": "7ca667fe4bb0ac74c0cb19f4a180d52fbe8dda920f0ba35a224f7ac6b7f44589",
+            "category": "queries"
+          },
+          "e-queries-011": {
+            "sha256": "ac0b01ee3be8f93dbccba584ac974756177618d59b71b46c2fc27f6ff369f901",
+            "category": "queries"
+          },
+          "e-queries-012": {
+            "sha256": "d65a677d03c6b58ceba2bbb63b8835c587a58f57772a17147a885ffb2ac5d0f9",
+            "category": "queries"
+          },
+          "e-queries-013": {
+            "sha256": "1178cf0ea0fe1c2a3733197707d54403130ca0ae73297e8c70752652255e3fdc",
+            "category": "queries"
+          },
+          "e-queries-014": {
+            "sha256": "1ee2b028a8e262af90b165ea479075352ea1c6ef8c48b72afbada4627d02a01e",
+            "category": "queries"
+          },
+          "e-queries-015": {
+            "sha256": "7875fad01cee86dbbdac9a719689675f8dd3d167a2c68ad090df1c373d10922f",
+            "category": "queries"
+          },
+          "e-queries-016": {
+            "sha256": "3d6140e0a78dcc8e90c330af84115befa0c9e0fa1b523458d64d919854a77371",
+            "category": "queries"
+          },
+          "e-queries-017": {
+            "sha256": "052483fee2ef13adbad2b973bd87527116c9b6fc7cc022480ad865f5476a0583",
+            "category": "queries"
+          },
+          "e-queries-018": {
+            "sha256": "dab7fe80b114e044be14736674519830dcebacf6e88411eec1788c01a280dd6e",
+            "category": "queries"
+          },
+          "e-queries-019": {
+            "sha256": "f333263e10c27578ebac3299a87748a2db881642214623766230b7f624e5d1a3",
+            "category": "queries"
+          },
+          "e-rest-api-001": {
+            "sha256": "d9ab7281153d491f55d242812a50e484ac2155888460978e009356dc5d0ba6ce",
+            "category": "rest-api"
+          },
+          "e-rest-api-002": {
+            "sha256": "678fad7743c4383f96042e2da3c30b5bb0ee07c9c4d39da9535e01e8bd16c775",
+            "category": "rest-api"
+          },
+          "e-rest-api-003": {
+            "sha256": "acf6a6fbbd88124e3518f81da30f422adea5e5405d6629189981e2f36fe4056d",
+            "category": "rest-api"
+          },
+          "e-rest-api-004": {
+            "sha256": "1b85069a8c618abc32ead24e0bb8c38be98ce7430808d051247f289918b7f6e8",
+            "category": "rest-api"
+          },
+          "e-rest-api-005": {
+            "sha256": "cfdfe1a8f821cda07290df45f7e0211117ef2c4b1701cf5275cd2806ce403105",
+            "category": "rest-api"
+          },
+          "e-rest-api-006": {
+            "sha256": "b13c89de43f7bb07696f916d06006a4d6fafdc6fb7bc82c2aa262a2ebce6c9c4",
+            "category": "rest-api"
+          },
+          "e-rest-api-007": {
+            "sha256": "94aa4df70916f8e031a7332766e7b13da4121130d16b0718c69fed464b8c6fb0",
+            "category": "rest-api"
+          },
+          "e-rest-api-008": {
+            "sha256": "6205bab9df485c5d56c4a088571e1fd227a32927ed798a472e4e0625e8ba4b96",
+            "category": "rest-api"
+          },
+          "e-rest-api-009": {
+            "sha256": "fa0741fc19b9506c31be12b677d232b8a9d95ab3e9b01df2e866f7d9513e7f94",
+            "category": "rest-api"
+          },
+          "e-rest-api-010": {
+            "sha256": "75a5f72ce329b45bb10bc67103b9a50f671e129fa40154851858c2b27a5d78d4",
+            "category": "rest-api"
+          },
+          "e-rest-api-011": {
+            "sha256": "e9a2be427963b1946ebfb0551c244d2c0e77e45c8efc7a6be26d9a5f9ad372e9",
+            "category": "rest-api"
+          },
+          "e-rest-api-012": {
+            "sha256": "40d2e9cca4197e66ea651eda4d3c551ad7fd1ab2b64d91d5d00e1b1d56b35148",
+            "category": "rest-api"
+          },
+          "e-rest-api-013": {
+            "sha256": "be8290358600946c4b7bc7d36b8b35489babb289ad95d7a5d735e39de8eb48c0",
+            "category": "rest-api"
+          },
+          "e-rest-api-014": {
+            "sha256": "89cdc01d63ed44a5b510d119be7aaf7fe6d98968a834181d26ef3eca8d10cf2d",
+            "category": "rest-api"
+          },
+          "e-rest-api-015": {
+            "sha256": "3dac803bcfad05c094e4f660881d53cf559a2738bf44b33a4c5a2fc427bb9f15",
+            "category": "rest-api"
+          },
+          "e-rest-api-016": {
+            "sha256": "66c2c8c612a840e90ea8ce963ecfda7a157f22d8275a662fa18cc9d78b6bae3c",
+            "category": "rest-api"
+          },
+          "e-rest-api-017": {
+            "sha256": "39373a52cd01fabcf4a44079cd4633e095e746f455b375c1dd47582ea1ebf991",
+            "category": "rest-api"
+          },
+          "e-rest-api-018": {
+            "sha256": "f068b574b3acb0aba5d976dd2156d7bf5543c2111476f507dd3beac5d640f0bf",
+            "category": "rest-api"
+          },
+          "e-rest-api-019": {
+            "sha256": "c18b595ee04bf69354252b9f759999d7e5883cc5017be77d38ff656c1cf4e9dc",
+            "category": "rest-api"
+          },
+          "e-rest-api-020": {
+            "sha256": "6931931c82563174c7eff4de42249ec8488ad4e8c4c389fabdcc711ebd2cde94",
+            "category": "rest-api"
+          },
+          "e-rest-api-021": {
+            "sha256": "306d551d11b6cb8707b2aa454670d3d371f86ddce54dd580d9da77b7d433546c",
+            "category": "rest-api"
+          },
+          "e-rest-api-022": {
+            "sha256": "77dd3a9b379e8c3e581cd8f089812dfb7c774ec1ce3607340d6eaeecf6ed4465",
+            "category": "rest-api"
+          },
+          "e-rest-api-023": {
+            "sha256": "9e65be64350f1d0b2b200f6c1c87154fee3666e44b2cc4641797ab85b965886b",
+            "category": "rest-api"
+          },
+          "e-rest-api-024": {
+            "sha256": "5f334566cd588cc64e98ad582a6c53c7328f58cdfa898a435bf30948205bc75c",
+            "category": "rest-api"
+          },
+          "e-rewrite-permalinks-001": {
+            "sha256": "967e5d11339c294adcb1b138baeedd15db74c638f28a7a2b355d28ab179dbc8c",
+            "category": "rewrite-permalinks"
+          },
+          "e-rewrite-permalinks-002": {
+            "sha256": "00493c6b5c4b04476faa710856a23e078225b4ef3ccb0bc2e11e51dae8ff402e",
+            "category": "rewrite-permalinks"
+          },
+          "e-rewrite-permalinks-003": {
+            "sha256": "a22ca57706ccc418f27f6a8aa8d7b4760a9837afb62e900f884b8f94a4663e4c",
+            "category": "rewrite-permalinks"
+          },
+          "e-rewrite-permalinks-004": {
+            "sha256": "12594a5f3f92a120ccff4d7644d968e985d8052497ad19d4b2abf5632066a89a",
+            "category": "rewrite-permalinks"
+          },
+          "e-rewrite-permalinks-005": {
+            "sha256": "539fa30daaaa0e97d4787d9a7507ce666b426370d2b614346dc8272fb7b83fd4",
+            "category": "rewrite-permalinks"
+          },
+          "e-rewrite-permalinks-006": {
+            "sha256": "e6bbb4bbf43c9736838ad28a51a294b260a62189af3fe7b25c998a558b878887",
+            "category": "rewrite-permalinks"
+          },
+          "e-roles-caps-001": {
+            "sha256": "ce22427aea55b9b8d506345bf6cb8ef55be98353962dc18336353acc2f415f38",
+            "category": "roles-caps"
+          },
+          "e-roles-caps-002": {
+            "sha256": "f3894de0a9e966015bf62e561ee21fdc6201574fee789540bc3595ccd32925d2",
+            "category": "roles-caps"
+          },
+          "e-roles-caps-003": {
+            "sha256": "96c78acc40f95f56906ec8d1aac3badafd2e48765f9e20bd1901f7ecdd3fc18a",
+            "category": "roles-caps"
+          },
+          "e-roles-caps-004": {
+            "sha256": "12f4243f1a1b36c994e8bbb145b0d3fb89f7c1dd92fb3680f8a3fedc82024218",
+            "category": "roles-caps"
+          },
+          "e-roles-caps-005": {
+            "sha256": "68d1db17fe8392e8f24aeb6b9f05040d958441f6d6479b5bcf5a454c301f2704",
+            "category": "roles-caps"
+          },
+          "e-roles-caps-006": {
+            "sha256": "56b90d0703a89189b4431f2f9e9defff28622b41d3cc00ca7817cc8a545ab5f3",
+            "category": "roles-caps"
+          },
+          "e-roles-caps-007": {
+            "sha256": "b1f63108faa5014bb1c5038c1574bef3b24909df63a4cd5c342d9dc5d5c27dea",
+            "category": "roles-caps"
+          },
+          "e-roles-caps-008": {
+            "sha256": "bba927d3ab0c8fd72cba425337880abe168e55a5e3175d25196902fe273dc810",
+            "category": "roles-caps"
+          },
+          "e-roles-caps-009": {
+            "sha256": "03b876ca6cf578cd1311b6e308cba5b0a3483ca6bd2beb917679c6b544838ae6",
+            "category": "roles-caps"
+          },
+          "e-roles-caps-010": {
+            "sha256": "edf2f8034f77a01d5c2783312b2ce17d71481e992bc5360d3534728ed5943801",
+            "category": "roles-caps"
+          },
+          "e-scripts-styles-001": {
+            "sha256": "02d1a6a3acbf61bbb13e7b2585fdc8300af480da163457a4f138ad8d07cf2a31",
+            "category": "scripts-styles"
+          },
+          "e-scripts-styles-002": {
+            "sha256": "f6dcd1bb1bf6fce42faaecec76f7830efa4b73309deebf69285bca11c6f4fd5b",
+            "category": "scripts-styles"
+          },
+          "e-scripts-styles-003": {
+            "sha256": "635392e52f1808bf220d3983f9ae1bbcccd2600727f56b2ab6a6aa832f489a4d",
+            "category": "scripts-styles"
+          },
+          "e-scripts-styles-004": {
+            "sha256": "b7dee4b4d1c733fd664ac95c0655e1387c0234aff0dfbc8eab70ec2ce4451dc9",
+            "category": "scripts-styles"
+          },
+          "e-scripts-styles-005": {
+            "sha256": "46bb678738ef7d67117c2ffb50b855ffd45656bdc161c429fac7a2a52729a6e0",
+            "category": "scripts-styles"
+          },
+          "e-security-001": {
+            "sha256": "285de0012b04f21480aa2051ccf887e94e1c47984ee02373319f7deaa8ed7914",
+            "category": "security"
+          },
+          "e-security-002": {
+            "sha256": "f9636bdb437b24ab2362a006e65f6d059b989debcd6e145998dc432fbe5f376f",
+            "category": "security"
+          },
+          "e-security-003": {
+            "sha256": "edaec4580f5d7a34e733ca6a41b3ccd0c19a587cfdf07c34a96ed4ccddc678ec",
+            "category": "security"
+          },
+          "e-security-004": {
+            "sha256": "414d6048edc7c00d6ff9c5bc4d17dd4fc31afdd55ca12a5fa0d31d73b4ad3564",
+            "category": "security"
+          },
+          "e-security-005": {
+            "sha256": "5b922b78eb03e1ca575c67cc96cd6333d9da74e7df92c13c6894286631a9f822",
+            "category": "security"
+          },
+          "e-security-006": {
+            "sha256": "e6a92248230e78c500ef2f712c020a7191d3d7dbe951378070884b554b875113",
+            "category": "security"
+          },
+          "e-security-007": {
+            "sha256": "123416ad09b0dbf6b225116f9d66dc8a9a180b774ad44d90dbddafc0027ebd71",
+            "category": "security"
+          },
+          "e-security-008": {
+            "sha256": "52acb723640e4657bbbc06c1462ccf1a717cb84bba3262a91d6c6f979c55a6ff",
+            "category": "security"
+          },
+          "e-security-009": {
+            "sha256": "2bda33ac1a5e942a599ad3d3439c5182280e5d41f525ca51ed2bf56dfe95489a",
+            "category": "security"
+          },
+          "e-security-010": {
+            "sha256": "3d53dc1d5c95981893bb8e2f0aaf7b39ed399fc5349c1e3e92e4378df4895e12",
+            "category": "security"
+          },
+          "e-security-011": {
+            "sha256": "e0c1e469f0b9415539dd9c560e83353222cb1e3d0ea35d8a1ed825866aa6bf22",
+            "category": "security"
+          },
+          "e-security-012": {
+            "sha256": "2926fdae0890db235fdecf3ce8d87d848900b4692e11e4868de9e9b5e9ef5073",
+            "category": "security"
+          },
+          "e-settings-options-001": {
+            "sha256": "681b37ae6f71f58b765807677b54402632e9db08e349a396cd69b6f7c43b4abf",
+            "category": "settings-options"
+          },
+          "e-settings-options-002": {
+            "sha256": "6f9c271418f996fa73090baf334af94a2d94ca2486bc45e463edc3a20c1dac9c",
+            "category": "settings-options"
+          },
+          "e-settings-options-003": {
+            "sha256": "ddf187cd0c4f2194ac09c32e5e47dbe46c35f027e2bfb490d6dc20367d2e67a4",
+            "category": "settings-options"
+          },
+          "e-settings-options-004": {
+            "sha256": "eacce56a125c4a843c1eff2739bcecdef9c9d1b67a64c344ae376c339ce8f250",
+            "category": "settings-options"
+          },
+          "e-settings-options-005": {
+            "sha256": "776ee219e8ec63bdd16942cd1443fe9c528fe7cb6dac797b16e6182169abb6be",
+            "category": "settings-options"
+          },
+          "e-settings-options-006": {
+            "sha256": "76ddf34e691490a5fa9e2130794811f6aa086f35b8a1ea8cf4e6476cfebb10e3",
+            "category": "settings-options"
+          },
+          "e-settings-options-007": {
+            "sha256": "60df513615bb66508df1f012cbccfdcace3bc0a7de02433623c67dbfdd532dcd",
+            "category": "settings-options"
+          },
+          "e-settings-options-008": {
+            "sha256": "3118ed256024157f2b6e0b9f01d58c5329f5a4e4833f3a8c8ee1e8955ee83cf1",
+            "category": "settings-options"
+          },
+          "e-settings-options-009": {
+            "sha256": "2a344efe63a2f37169dc0c828f1a69afedde1d3d41f6d82af315a76c0f898be9",
+            "category": "settings-options"
+          },
+          "e-settings-options-010": {
+            "sha256": "30bb54deb2fc8ec09d35099c2da387e1689235d392458807f955e2e005cd51d3",
+            "category": "settings-options"
+          },
+          "e-shortcodes-001": {
+            "sha256": "963c6a5792a9328d3b43c0992d6cde74110ecb518900b0ab1d8030023a8ae5ca",
+            "category": "shortcodes"
+          },
+          "e-shortcodes-002": {
+            "sha256": "954f03cd431638b955572c7cfe8c1614c977755870668cbfda9c17226acf06e9",
+            "category": "shortcodes"
+          },
+          "e-shortcodes-003": {
+            "sha256": "9826d11f94d99da14778eee80277a3ed8fb1243246f72e01de2cfd4e0cfc9a8d",
+            "category": "shortcodes"
+          },
+          "e-shortcodes-004": {
+            "sha256": "6c17c64b8a6466079412ab1be477be29c561cc4ebc5d51b0a201138b193582af",
+            "category": "shortcodes"
+          },
+          "e-shortcodes-005": {
+            "sha256": "3beb0bee322c5588898cad5fc93478152251e0eb4afac41558cbf4aa8f36abb1",
+            "category": "shortcodes"
+          },
+          "e-theme-templates-001": {
+            "sha256": "f8527a5b5382c9e43cca62f58b39b34a514138c84754b5157b49c492456b9aa6",
+            "category": "theme-templates"
+          },
+          "e-theme-templates-002": {
+            "sha256": "8b99b0d9833a35fcd6cb56cf3f3358dc1a4674889cb20073d57d22f28419d1bc",
+            "category": "theme-templates"
+          },
+          "e-theme-templates-003": {
+            "sha256": "bff6717888e27dc7fc3cee96747d4bbb8a174d3d1adbb0a0ee2b5a48b5e690f5",
+            "category": "theme-templates"
+          },
+          "e-theme-templates-004": {
+            "sha256": "294179329405945a06017cf043a4b267c7c19d3b99fd34f4cfdd321a90938a17",
+            "category": "theme-templates"
+          },
+          "e-theme-templates-005": {
+            "sha256": "688904294299c05b1c24522aca8f234c76a19284f2c5c3e3ba94217d5b2ad17b",
+            "category": "theme-templates"
+          },
+          "e-theme-templates-006": {
+            "sha256": "0027fb8c80a348315af6de675103d803d0759641808dde54c3409aa254d488e0",
+            "category": "theme-templates"
+          },
+          "e-theme-templates-007": {
+            "sha256": "ffb04ff7f8e838a0ce015f32c55c49ac2ed29037d5c4ebb180620d5b784fe7d7",
+            "category": "theme-templates"
+          },
+          "e-theme-templates-008": {
+            "sha256": "0c4ba4bb9a755a79f32564671902a3e6d25cf2ed8682406fe6713365931647f0",
+            "category": "theme-templates"
+          },
+          "e-theme-templates-009": {
+            "sha256": "d8ecf1b85393292a7b1d03eeae6f333125874d2046e42f0f00c7ef47e0f5b0a3",
+            "category": "theme-templates"
+          },
+          "e-theme-templates-010": {
+            "sha256": "37af351e0d0e19e7973c019f247ccade7b757ccb2848f0b1217763155ca7887b",
+            "category": "theme-templates"
+          },
+          "e-theme-templates-011": {
+            "sha256": "bfa08ffa6a37632638c43c6a653010f8fd458c5bdea1d64e09f6e0d160e78080",
+            "category": "theme-templates"
+          }
+        }
+      },
+      "wp-cli": {
+        "task_count": 27,
+        "tasks": {
+          "e-cli-content-002": {
+            "sha256": "ce2add7db4e00b84ffddadcb45d67947d7de1bc0844a4b9b3b34b8a058b213a4",
+            "category": "cli-content"
+          },
+          "e-cli-content-003": {
+            "sha256": "6416b2553ce430bf69bea9776cfe252fc2a328727c873f0d5d2f98b35725014e",
+            "category": "cli-content"
+          },
+          "e-cli-content-004": {
+            "sha256": "8213ace8a9fdf1476f8c713c65e7d805f0096402849cf20b878725abe8aea3ec",
+            "category": "cli-content"
+          },
+          "e-cli-maintenance-001": {
+            "sha256": "8d37474d528f1164fd360d61ba45f06ebd215eb959cddc0efc93bc455d3812a0",
+            "category": "cli-maintenance"
+          },
+          "e-cli-maintenance-002": {
+            "sha256": "b8d0878edfeec1b99d813f6e420e3c6ab059690fe88c28c85d275509f9d83c92",
+            "category": "cli-maintenance"
+          },
+          "e-cli-maintenance-003": {
+            "sha256": "f2a799c438434c1ddec3f9d98222d1c17420fcbae82d8de37b8d3f9fdfa40949",
+            "category": "cli-maintenance"
+          },
+          "e-cli-maintenance-004": {
+            "sha256": "c01e091c69db679099543b7c65e059d72ec960bae5cf25b9a489d1e630b0a5ed",
+            "category": "cli-maintenance"
+          },
+          "e-cli-maintenance-005": {
+            "sha256": "ac0fb3d555ff811ae334c8ac31cd044b52ca825072a4ac4ef826c0e149b50e63",
+            "category": "cli-maintenance"
+          },
+          "e-cli-maintenance-006": {
+            "sha256": "dff9bf6f0c132fc19cb333f8a84058efda41cf39279c0521514ce8eafcc1c5ee",
+            "category": "cli-maintenance"
+          },
+          "e-cli-migration-001": {
+            "sha256": "07381136df2708db5193dc44e6cad520c711f5f83845cdd3acf04c421dec657a",
+            "category": "cli-migration"
+          },
+          "e-cli-migration-002": {
+            "sha256": "9bcaf8f60dde80d8e01e26c1822450a126056486c411159e22bbfb8aa368b6f6",
+            "category": "cli-migration"
+          },
+          "e-cli-options-001": {
+            "sha256": "8070e8f55e35e9f526ba029f380db7ef65b6b389300aba24702ac8c904760809",
+            "category": "cli-options"
+          },
+          "e-cli-options-003": {
+            "sha256": "5a2c65a56a8c0adecf6b1e0701d64b2819edb279093a9b0521ca4b89bf633996",
+            "category": "cli-options"
+          },
+          "e-cli-options-004": {
+            "sha256": "25e34b1d0a08d1d4ce72d53a0619b0850502ca305326b93f8536ba582af6082c",
+            "category": "cli-options"
+          },
+          "e-cli-options-005": {
+            "sha256": "d8ceb48dcfe168a620cc9790fc2ae3e80cfea1f6ea33fba2d52da12c2dc1468a",
+            "category": "cli-options"
+          },
+          "e-cli-reporting-001": {
+            "sha256": "a033c28e62bd407bd58728f9c035012dcea4dd3d6c9dd8224f59ce4b31f8ad7d",
+            "category": "cli-reporting"
+          },
+          "e-cli-reporting-002": {
+            "sha256": "4ec28641a37806b31463352fc4572d3d7fd3c30894fb682eedb56c1a0c863935",
+            "category": "cli-reporting"
+          },
+          "e-cli-reporting-004": {
+            "sha256": "05262f1ab6cb1e75a0790c175be00f016bb3e7514ef5a800405128ba0f2928ed",
+            "category": "cli-reporting"
+          },
+          "e-cli-taxonomy-meta-001": {
+            "sha256": "38f02e100479909fce654594550d09a5975bc307c8e8ffe0b2ab009edba12d67",
+            "category": "cli-taxonomy-meta"
+          },
+          "e-cli-taxonomy-meta-002": {
+            "sha256": "bc8489f9891de385bcfd47e12af2cbb1e44f965df57d557438e2912485932de7",
+            "category": "cli-taxonomy-meta"
+          },
+          "e-cli-taxonomy-meta-004": {
+            "sha256": "0dee15444c2719cce56539b0f3af59b6ab5bef73683ed39919ca5b3773ce3ac5",
+            "category": "cli-taxonomy-meta"
+          },
+          "e-cli-taxonomy-meta-005": {
+            "sha256": "b23366af2c9fb965e16435d7980f34b5b67f0773252094d1339cc4760e2b978a",
+            "category": "cli-taxonomy-meta"
+          },
+          "e-cli-users-002": {
+            "sha256": "14b54832a4d080b0fb08c040a083d376eb310e2cf8c7c6ea5019e6ff88a0d491",
+            "category": "cli-users"
+          },
+          "e-cli-users-003": {
+            "sha256": "b5cfd991e7e1c9a2c1dfe37fc86c434a85e1d5f85bfdf4d6939b22cab6d471c6",
+            "category": "cli-users"
+          },
+          "e-cli-users-004": {
+            "sha256": "2924923d621a991251f8a7d3edf1a7823ddd66df41602e161065e55745dfedaf",
+            "category": "cli-users"
+          },
+          "e-cli-users-005": {
+            "sha256": "b6496d115bee0f083ce2fbe9983b64bdb12f7449ee90d8a7de530fbdc13b6dc5",
+            "category": "cli-users"
+          },
+          "e-cli-users-006": {
+            "sha256": "caf1a3204c54b3d88491495ad99c3874265b4ea114969ea5555fb33009914055",
+            "category": "cli-users"
+          }
+        }
+      }
+    }
+  }
+}
diff --git a/datasets/README.md b/datasets/README.md
index 61d59d7..9018d11 100644
--- a/datasets/README.md
+++ b/datasets/README.md
@@ -7,12 +7,12 @@ This directory contains the benchmark test suites and tooling for publishing to
 ```
 datasets/
 ├── suites/                    # Source of truth (human-editable JSON)
-│   ├── wp-core-v1/
+│   ├── wp-core/
 │   │   └── execution/         # Code generation tests (one file per category)
 │   │       ├── hooks.json
 │   │       ├── rest-api.json
 │   │       └── ...
-│   └── wp-cli-v1/
+│   └── wp-cli/
 │       └── execution/         # 27 WP-CLI workflows, seven categories
 ├── data/                      # Generated Parquet for HF (gitignored)
 │   └── test.parquet
@@ -28,7 +28,7 @@ The harness loads directly from `suites/` JSON files:
 # wp-bench.yaml
 dataset:
   source: local
-  name: wp-core-v1
+  name: wp-core
 ```
 
 ## Publishing to Hugging Face
@@ -108,7 +108,7 @@ The local loader selects `dataset.name`; the Hub loader uses `dataset.name` as
 the repository and `dataset.suite` to filter the combined export, defaulting to
 `run.suite`. An explicit Hub `dataset.suite` also sets the result-suite label.
 `--suite` handles both sources. Exports include both suites, and omit authored
-exploit candidates. CLI scores should be reported separately from core scores.
+exploit candidates. Report component/category scores alongside the task-weighted overall score defined by the public benchmark release.
 
 
 ### Per-Test Metadata
@@ -132,3 +132,11 @@ field named `suite_metadata` would be shadowed — avoid that name).
 
 All fields are optional; unknown fields are preserved as-is. Prefer adding
 new fields here before inventing per-suite conventions.
+
+## Release identity
+
+`wp-core` and `wp-cli` are stable component IDs. Category JSON files carry no
+independent version. The complete public release and all source fingerprints
+are defined in [../benchmark/README.md](../benchmark/README.md). Any dataset
+change creates an unreleased evaluation until the next overall release is frozen.
+Exports use the stable IDs; publish a pinned Hub revision for each public release.
diff --git a/datasets/suites/wp-cli-v1/execution/cli-content.json b/datasets/suites/wp-cli/execution/cli-content.json
similarity index 95%
rename from datasets/suites/wp-cli-v1/execution/cli-content.json
rename to datasets/suites/wp-cli/execution/cli-content.json
index d8e58cc..673cbfe 100644
--- a/datasets/suites/wp-cli-v1/execution/cli-content.json
+++ b/datasets/suites/wp-cli/execution/cli-content.json
@@ -1,5 +1,5 @@
 {
-  "id": "wp-cli-v1",
+  "id": "wp-cli",
   "name": "WP-CLI operational workflows: cli-content",
   "metadata": {
     "wp_version": "7.1",
@@ -24,8 +24,8 @@
       "static_checks": {
         "forbidden_patterns": [
           {
-            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
-            "description": "Use native WordPress operations; read-only table/prefix discovery is allowed, but PHP evaluation and direct SQL are forbidden",
+            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables|columns)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
+            "description": "Use native WordPress operations; read-only table, prefix, and column discovery is allowed, but PHP evaluation and direct SQL are forbidden",
             "severity": "error"
           }
         ]
@@ -82,8 +82,8 @@
       "static_checks": {
         "forbidden_patterns": [
           {
-            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
-            "description": "Use native WordPress operations; read-only table/prefix discovery is allowed, but PHP evaluation and direct SQL are forbidden",
+            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables|columns)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
+            "description": "Use native WordPress operations; read-only table, prefix, and column discovery is allowed, but PHP evaluation and direct SQL are forbidden",
             "severity": "error"
           }
         ]
@@ -139,8 +139,8 @@
       "static_checks": {
         "forbidden_patterns": [
           {
-            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
-            "description": "Use native WordPress operations; read-only table/prefix discovery is allowed, but PHP evaluation and direct SQL are forbidden",
+            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables|columns)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
+            "description": "Use native WordPress operations; read-only table, prefix, and column discovery is allowed, but PHP evaluation and direct SQL are forbidden",
             "severity": "error"
           }
         ]
diff --git a/datasets/suites/wp-cli-v1/execution/cli-maintenance.json b/datasets/suites/wp-cli/execution/cli-maintenance.json
similarity index 96%
rename from datasets/suites/wp-cli-v1/execution/cli-maintenance.json
rename to datasets/suites/wp-cli/execution/cli-maintenance.json
index 4aa388d..e99303c 100644
--- a/datasets/suites/wp-cli-v1/execution/cli-maintenance.json
+++ b/datasets/suites/wp-cli/execution/cli-maintenance.json
@@ -1,5 +1,5 @@
 {
-  "id": "wp-cli-v1",
+  "id": "wp-cli",
   "name": "WP-CLI operational workflows: cli-maintenance",
   "metadata": {
     "wp_version": "7.1",
@@ -24,8 +24,8 @@
       "static_checks": {
         "forbidden_patterns": [
           {
-            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
-            "description": "Use native WordPress operations; read-only table/prefix discovery is allowed, but PHP evaluation and direct SQL are forbidden",
+            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables|columns)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
+            "description": "Use native WordPress operations; read-only table, prefix, and column discovery is allowed, but PHP evaluation and direct SQL are forbidden",
             "severity": "error"
           }
         ]
@@ -81,8 +81,8 @@
       "static_checks": {
         "forbidden_patterns": [
           {
-            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
-            "description": "Use native WordPress operations; read-only table/prefix discovery is allowed, but PHP evaluation and direct SQL are forbidden",
+            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables|columns)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
+            "description": "Use native WordPress operations; read-only table, prefix, and column discovery is allowed, but PHP evaluation and direct SQL are forbidden",
             "severity": "error"
           }
         ]
@@ -137,8 +137,8 @@
       "static_checks": {
         "forbidden_patterns": [
           {
-            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
-            "description": "Use native WordPress operations; read-only table/prefix discovery is allowed, but PHP evaluation and direct SQL are forbidden",
+            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables|columns)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
+            "description": "Use native WordPress operations; read-only table, prefix, and column discovery is allowed, but PHP evaluation and direct SQL are forbidden",
             "severity": "error"
           }
         ]
@@ -196,8 +196,8 @@
       "static_checks": {
         "forbidden_patterns": [
           {
-            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
-            "description": "Use native WordPress operations; read-only table/prefix discovery is allowed, but PHP evaluation and direct SQL are forbidden",
+            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables|columns)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
+            "description": "Use native WordPress operations; read-only table, prefix, and column discovery is allowed, but PHP evaluation and direct SQL are forbidden",
             "severity": "error"
           }
         ]
@@ -258,8 +258,8 @@
       "static_checks": {
         "forbidden_patterns": [
           {
-            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
-            "description": "Use native WordPress operations; read-only table/prefix discovery is allowed, but PHP evaluation and direct SQL are forbidden",
+            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables|columns)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
+            "description": "Use native WordPress operations; read-only table, prefix, and column discovery is allowed, but PHP evaluation and direct SQL are forbidden",
             "severity": "error"
           }
         ]
@@ -317,8 +317,8 @@
       "static_checks": {
         "forbidden_patterns": [
           {
-            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
-            "description": "Use native WordPress operations; read-only table/prefix discovery is allowed, but PHP evaluation and direct SQL are forbidden",
+            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables|columns)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
+            "description": "Use native WordPress operations; read-only table, prefix, and column discovery is allowed, but PHP evaluation and direct SQL are forbidden",
             "severity": "error"
           }
         ]
diff --git a/datasets/suites/wp-cli-v1/execution/cli-migration.json b/datasets/suites/wp-cli/execution/cli-migration.json
similarity index 97%
rename from datasets/suites/wp-cli-v1/execution/cli-migration.json
rename to datasets/suites/wp-cli/execution/cli-migration.json
index eac7360..41848db 100644
--- a/datasets/suites/wp-cli-v1/execution/cli-migration.json
+++ b/datasets/suites/wp-cli/execution/cli-migration.json
@@ -1,5 +1,5 @@
 {
-  "id": "wp-cli-v1",
+  "id": "wp-cli",
   "name": "WP-CLI operational workflows: cli-migration",
   "metadata": {
     "wp_version": "7.1",
@@ -26,8 +26,8 @@
       "static_checks": {
         "forbidden_patterns": [
           {
-            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
-            "description": "Use native WordPress operations; read-only table/prefix discovery is allowed, but PHP evaluation and direct SQL are forbidden",
+            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables|columns)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
+            "description": "Use native WordPress operations; read-only table, prefix, and column discovery is allowed, but PHP evaluation and direct SQL are forbidden",
             "severity": "error"
           }
         ]
@@ -89,8 +89,8 @@
       "static_checks": {
         "forbidden_patterns": [
           {
-            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
-            "description": "Use native WordPress operations; read-only table/prefix discovery is allowed, but PHP evaluation and direct SQL are forbidden",
+            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables|columns)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
+            "description": "Use native WordPress operations; read-only table, prefix, and column discovery is allowed, but PHP evaluation and direct SQL are forbidden",
             "severity": "error"
           }
         ]
diff --git a/datasets/suites/wp-cli-v1/execution/cli-options.json b/datasets/suites/wp-cli/execution/cli-options.json
similarity index 95%
rename from datasets/suites/wp-cli-v1/execution/cli-options.json
rename to datasets/suites/wp-cli/execution/cli-options.json
index 8c23edd..fb318a0 100644
--- a/datasets/suites/wp-cli-v1/execution/cli-options.json
+++ b/datasets/suites/wp-cli/execution/cli-options.json
@@ -1,5 +1,5 @@
 {
-  "id": "wp-cli-v1",
+  "id": "wp-cli",
   "name": "WP-CLI operational workflows: cli-options",
   "metadata": {
     "wp_version": "7.1",
@@ -24,8 +24,8 @@
       "static_checks": {
         "forbidden_patterns": [
           {
-            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
-            "description": "Use native WordPress operations; read-only table/prefix discovery is allowed, but PHP evaluation and direct SQL are forbidden",
+            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables|columns)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
+            "description": "Use native WordPress operations; read-only table, prefix, and column discovery is allowed, but PHP evaluation and direct SQL are forbidden",
             "severity": "error"
           }
         ]
@@ -80,8 +80,8 @@
       "static_checks": {
         "forbidden_patterns": [
           {
-            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
-            "description": "Use native WordPress operations; read-only table/prefix discovery is allowed, but PHP evaluation and direct SQL are forbidden",
+            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables|columns)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
+            "description": "Use native WordPress operations; read-only table, prefix, and column discovery is allowed, but PHP evaluation and direct SQL are forbidden",
             "severity": "error"
           }
         ]
@@ -137,8 +137,8 @@
       "static_checks": {
         "forbidden_patterns": [
           {
-            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
-            "description": "Use native WordPress operations; read-only table/prefix discovery is allowed, but PHP evaluation and direct SQL are forbidden",
+            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables|columns)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
+            "description": "Use native WordPress operations; read-only table, prefix, and column discovery is allowed, but PHP evaluation and direct SQL are forbidden",
             "severity": "error"
           }
         ]
@@ -195,8 +195,8 @@
       "static_checks": {
         "forbidden_patterns": [
           {
-            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
-            "description": "Use native WordPress operations; read-only table/prefix discovery is allowed, but PHP evaluation and direct SQL are forbidden",
+            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables|columns)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
+            "description": "Use native WordPress operations; read-only table, prefix, and column discovery is allowed, but PHP evaluation and direct SQL are forbidden",
             "severity": "error"
           }
         ]
diff --git a/datasets/suites/wp-cli-v1/execution/cli-reporting.json b/datasets/suites/wp-cli/execution/cli-reporting.json
similarity index 96%
rename from datasets/suites/wp-cli-v1/execution/cli-reporting.json
rename to datasets/suites/wp-cli/execution/cli-reporting.json
index babf5c7..2fca47f 100644
--- a/datasets/suites/wp-cli-v1/execution/cli-reporting.json
+++ b/datasets/suites/wp-cli/execution/cli-reporting.json
@@ -1,5 +1,5 @@
 {
-  "id": "wp-cli-v1",
+  "id": "wp-cli",
   "name": "WP-CLI operational workflows: cli-reporting",
   "metadata": {
     "wp_version": "7.1",
@@ -24,8 +24,8 @@
       "static_checks": {
         "forbidden_patterns": [
           {
-            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
-            "description": "Use native WordPress operations; read-only table/prefix discovery is allowed, but PHP evaluation and direct SQL are forbidden",
+            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables|columns)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
+            "description": "Use native WordPress operations; read-only table, prefix, and column discovery is allowed, but PHP evaluation and direct SQL are forbidden",
             "severity": "error"
           }
         ]
@@ -89,8 +89,8 @@
       "static_checks": {
         "forbidden_patterns": [
           {
-            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
-            "description": "Use native WordPress operations; read-only table/prefix discovery is allowed, but PHP evaluation and direct SQL are forbidden",
+            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables|columns)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
+            "description": "Use native WordPress operations; read-only table, prefix, and column discovery is allowed, but PHP evaluation and direct SQL are forbidden",
             "severity": "error"
           }
         ]
@@ -146,8 +146,8 @@
       "static_checks": {
         "forbidden_patterns": [
           {
-            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
-            "description": "Use native WordPress operations; read-only table/prefix discovery is allowed, but PHP evaluation and direct SQL are forbidden",
+            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables|columns)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
+            "description": "Use native WordPress operations; read-only table, prefix, and column discovery is allowed, but PHP evaluation and direct SQL are forbidden",
             "severity": "error"
           }
         ]
diff --git a/datasets/suites/wp-cli-v1/execution/cli-taxonomy-meta.json b/datasets/suites/wp-cli/execution/cli-taxonomy-meta.json
similarity index 95%
rename from datasets/suites/wp-cli-v1/execution/cli-taxonomy-meta.json
rename to datasets/suites/wp-cli/execution/cli-taxonomy-meta.json
index 50bea88..903fa73 100644
--- a/datasets/suites/wp-cli-v1/execution/cli-taxonomy-meta.json
+++ b/datasets/suites/wp-cli/execution/cli-taxonomy-meta.json
@@ -1,5 +1,5 @@
 {
-  "id": "wp-cli-v1",
+  "id": "wp-cli",
   "name": "WP-CLI operational workflows: cli-taxonomy-meta",
   "metadata": {
     "wp_version": "7.1",
@@ -23,8 +23,8 @@
       "static_checks": {
         "forbidden_patterns": [
           {
-            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
-            "description": "Use native WordPress operations; read-only table/prefix discovery is allowed, but PHP evaluation and direct SQL are forbidden",
+            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables|columns)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
+            "description": "Use native WordPress operations; read-only table, prefix, and column discovery is allowed, but PHP evaluation and direct SQL are forbidden",
             "severity": "error"
           }
         ]
@@ -80,8 +80,8 @@
       "static_checks": {
         "forbidden_patterns": [
           {
-            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
-            "description": "Use native WordPress operations; read-only table/prefix discovery is allowed, but PHP evaluation and direct SQL are forbidden",
+            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables|columns)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
+            "description": "Use native WordPress operations; read-only table, prefix, and column discovery is allowed, but PHP evaluation and direct SQL are forbidden",
             "severity": "error"
           }
         ]
@@ -137,8 +137,8 @@
       "static_checks": {
         "forbidden_patterns": [
           {
-            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
-            "description": "Use native WordPress operations; read-only table/prefix discovery is allowed, but PHP evaluation and direct SQL are forbidden",
+            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables|columns)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
+            "description": "Use native WordPress operations; read-only table, prefix, and column discovery is allowed, but PHP evaluation and direct SQL are forbidden",
             "severity": "error"
           }
         ]
@@ -195,8 +195,8 @@
       "static_checks": {
         "forbidden_patterns": [
           {
-            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
-            "description": "Use native WordPress operations; read-only table/prefix discovery is allowed, but PHP evaluation and direct SQL are forbidden",
+            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables|columns)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
+            "description": "Use native WordPress operations; read-only table, prefix, and column discovery is allowed, but PHP evaluation and direct SQL are forbidden",
             "severity": "error"
           }
         ]
diff --git a/datasets/suites/wp-cli-v1/execution/cli-users.json b/datasets/suites/wp-cli/execution/cli-users.json
similarity index 96%
rename from datasets/suites/wp-cli-v1/execution/cli-users.json
rename to datasets/suites/wp-cli/execution/cli-users.json
index b7ee476..058af25 100644
--- a/datasets/suites/wp-cli-v1/execution/cli-users.json
+++ b/datasets/suites/wp-cli/execution/cli-users.json
@@ -1,5 +1,5 @@
 {
-  "id": "wp-cli-v1",
+  "id": "wp-cli",
   "name": "WP-CLI operational workflows: cli-users",
   "metadata": {
     "wp_version": "7.1",
@@ -23,8 +23,8 @@
       "static_checks": {
         "forbidden_patterns": [
           {
-            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
-            "description": "Use native WordPress operations; read-only table/prefix discovery is allowed, but PHP evaluation and direct SQL are forbidden",
+            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables|columns)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
+            "description": "Use native WordPress operations; read-only table, prefix, and column discovery is allowed, but PHP evaluation and direct SQL are forbidden",
             "severity": "error"
           }
         ]
@@ -86,8 +86,8 @@
       "static_checks": {
         "forbidden_patterns": [
           {
-            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
-            "description": "Use native WordPress operations; read-only table/prefix discovery is allowed, but PHP evaluation and direct SQL are forbidden",
+            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables|columns)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
+            "description": "Use native WordPress operations; read-only table, prefix, and column discovery is allowed, but PHP evaluation and direct SQL are forbidden",
             "severity": "error"
           }
         ]
@@ -143,8 +143,8 @@
       "static_checks": {
         "forbidden_patterns": [
           {
-            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
-            "description": "Use native WordPress operations; read-only table/prefix discovery is allowed, but PHP evaluation and direct SQL are forbidden",
+            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables|columns)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
+            "description": "Use native WordPress operations; read-only table, prefix, and column discovery is allowed, but PHP evaluation and direct SQL are forbidden",
             "severity": "error"
           }
         ]
@@ -201,8 +201,8 @@
       "static_checks": {
         "forbidden_patterns": [
           {
-            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
-            "description": "Use native WordPress operations; read-only table/prefix discovery is allowed, but PHP evaluation and direct SQL are forbidden",
+            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables|columns)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
+            "description": "Use native WordPress operations; read-only table, prefix, and column discovery is allowed, but PHP evaluation and direct SQL are forbidden",
             "severity": "error"
           }
         ]
@@ -262,8 +262,8 @@
       "static_checks": {
         "forbidden_patterns": [
           {
-            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
-            "description": "Use native WordPress operations; read-only table/prefix discovery is allowed, but PHP evaluation and direct SQL are forbidden",
+            "pattern": "/(?:\\bwp\\s+(?:--[^\\s]+\\s+)*(?:eval(?:-file)?|db(?!\\s+(?:--[^\\s]+\\s+)*(?:prefix|tables|columns)\\b)|shell)\\b|--(?:exec|require)(?:=|\\s)|(?:^|[\\s;&|])(?:php|sqlite3|mysql)\\s)/m",
+            "description": "Use native WordPress operations; read-only table, prefix, and column discovery is allowed, but PHP evaluation and direct SQL are forbidden",
             "severity": "error"
           }
         ]
diff --git a/datasets/suites/wp-core-v1/execution/abilities-api.json b/datasets/suites/wp-core/execution/abilities-api.json
similarity index 97%
rename from datasets/suites/wp-core-v1/execution/abilities-api.json
rename to datasets/suites/wp-core/execution/abilities-api.json
index 56c9339..f2b5c89 100644
--- a/datasets/suites/wp-core-v1/execution/abilities-api.json
+++ b/datasets/suites/wp-core/execution/abilities-api.json
@@ -1,6 +1,5 @@
 {
-  "id": "wp-core-execution-v1-abilities_api",
-  "version": "2.3.0",
+  "id": "wp-core",
   "metadata": {
     "name": "WordPress Core Execution Tests - Abilities API",
     "description": "Code generation tasks for WordPress Abilities API APIs",
@@ -182,7 +181,7 @@
       "id": "e-abilities-api-004",
       "category": "abilities-api",
       "prompt": "Register an Abilities API ability named 'wpbp/rest-visible' in the 'wpbp-rest-abilities' category and configure the ability to be available through the Abilities REST API.",
-      "expected_behavior": "Registers the `wpbp-rest-abilities` category and the `wpbp/rest-visible` ability through the Abilities API init hooks, with `meta.show_in_rest` set to true on the ability. Runtime validation inspects the registered ability metadata and executes its callback; REST visibility should come from Abilities API metadata, not a custom REST route.",
+      "expected_behavior": "The ability is registered in the requested category with REST visibility metadata. Its callback result is unconstrained because the prompt does not specify a return payload.",
       "requirements": [
         "Use WordPress Abilities API",
         "Configure the ability for Abilities REST API availability",
@@ -233,7 +232,7 @@
         "assertions": [
           {
             "type": "custom_assertion",
-            "code": "WP_Abilities_Registry::get_instance(); $ability = wp_get_ability( 'wpbp/rest-visible' ); if ( ! $ability instanceof WP_Ability ) { return false; } $meta = $ability->get_meta(); return 'wpbp-rest-abilities' === $ability->get_category() && true === ( $meta['show_in_rest'] ?? false ) && array( 'ok' => true ) === $ability->execute();",
+            "code": "WP_Abilities_Registry::get_instance(); $ability = wp_get_ability( 'wpbp/rest-visible' ); if ( ! $ability instanceof WP_Ability ) { return false; } $meta = $ability->get_meta(); return 'wpbp-rest-abilities' === $ability->get_category() && true === ( $meta['show_in_rest'] ?? false );",
             "description": "The ability is registered with REST visibility metadata",
             "weight": 1
           }
@@ -248,7 +247,11 @@
           "src/wp-includes/abilities-api/class-wp-ability-categories-registry.php"
         ],
         "release_focus": "6.9"
-      }
+      },
+      "exploit_solutions": [
+        "add_action( 'wp_abilities_api_categories_init', function () { wp_register_ability_category( 'wpbp-rest-abilities', array( 'label' => 'REST', 'description' => 'REST abilities.' ) ); } ); add_action( 'wp_abilities_api_init', function () { wp_register_ability( 'wpbp/rest-visible', array( 'label' => 'Probe', 'description' => 'Probe', 'category' => 'wpbp-rest-abilities', 'execute_callback' => '__return_true', 'permission_callback' => '__return_true' ) ); } );",
+        "add_action( 'wp_abilities_api_categories_init', function () { wp_register_ability_category( 'wpbp-wrong', array( 'label' => 'Wrong', 'description' => 'Wrong category.' ) ); } ); add_action( 'wp_abilities_api_init', function () { wp_register_ability( 'wpbp/rest-visible', array( 'label' => 'Probe', 'description' => 'Probe', 'category' => 'wpbp-wrong', 'execute_callback' => '__return_true', 'permission_callback' => '__return_true', 'meta' => array( 'show_in_rest' => true ) ) ); } );"
+      ]
     },
     {
       "id": "e-abilities-api-005",
diff --git a/datasets/suites/wp-core-v1/execution/admin.json b/datasets/suites/wp-core/execution/admin.json
similarity index 99%
rename from datasets/suites/wp-core-v1/execution/admin.json
rename to datasets/suites/wp-core/execution/admin.json
index 2bb2f42..0c26ba1 100644
--- a/datasets/suites/wp-core-v1/execution/admin.json
+++ b/datasets/suites/wp-core/execution/admin.json
@@ -1,6 +1,5 @@
 {
-  "id": "wp-core-execution-v1-admin",
-  "version": "1.0.1",
+  "id": "wp-core",
   "metadata": {
     "name": "WordPress Core Execution Tests - Admin",
     "description": "Code generation tasks for the WordPress admin: menus, list tables, settings, AJAX and notices",
diff --git a/datasets/suites/wp-core-v1/execution/ai-client.json b/datasets/suites/wp-core/execution/ai-client.json
similarity index 93%
rename from datasets/suites/wp-core-v1/execution/ai-client.json
rename to datasets/suites/wp-core/execution/ai-client.json
index 79295ba..b5bbdf0 100644
--- a/datasets/suites/wp-core-v1/execution/ai-client.json
+++ b/datasets/suites/wp-core/execution/ai-client.json
@@ -1,6 +1,5 @@
 {
-  "id": "wp-core-execution-v1-ai_client",
-  "version": "2.4.0",
+  "id": "wp-core",
   "metadata": {
     "name": "WordPress Core Execution Tests - AI Client",
     "description": "Code generation tasks for WordPress AI Client APIs",
@@ -60,7 +59,7 @@
       "id": "e-ai-client-002",
       "category": "ai-client",
       "prompt": "Implement a function that creates a text prompt builder with a concise system instruction, temperature 0.2, and a model preference order of 'gpt-4o-mini' then 'gpt-4o'. Return the builder without generating content.",
-      "expected_behavior": "`wpbp_ai_client_002()` returns a `WP_AI_Client_Prompt_Builder` from `wp_ai_client_prompt()` after configuring text-generation options with snake_case AI Client methods. It must not call a generating method; runtime validation later prevents generation and confirms the returned builder is not already in an error state.",
+      "expected_behavior": "The returned builder must contain the requested temperature, a nonempty system instruction, and the requested model preference order, and must remain ready for a later prevented generation. Runtime inspection uses the pinned WordPress 7.1 SDK fields without making a provider call.",
       "requirements": [
         "Use wp_ai_client_prompt()",
         "Configure temperature, system instruction, and model preference",
@@ -102,7 +101,7 @@
         "assertions": [
           {
             "type": "custom_assertion",
-            "code": "if ( ! function_exists( 'wpbp_ai_client_002' ) ) { return false; } $builder = wpbp_ai_client_002(); if ( ! $builder instanceof WP_AI_Client_Prompt_Builder ) { return false; } add_filter( 'wp_ai_client_prevent_prompt', '__return_true' ); $result = $builder->generate_text(); remove_filter( 'wp_ai_client_prevent_prompt', '__return_true' ); return is_wp_error( $result ) && 'prompt_prevented' === $result->get_error_code();",
+            "code": "if ( ! function_exists( 'wpbp_ai_client_002' ) ) { return false; } $builder = wpbp_ai_client_002(); if ( ! $builder instanceof WP_AI_Client_Prompt_Builder ) { return false; } $wrapped = ( new ReflectionProperty( WP_AI_Client_Prompt_Builder::class, 'builder' ) )->getValue( $builder ); $config = ( new ReflectionProperty( $wrapped, 'modelConfig' ) )->getValue( $wrapped ); $preferences = ( new ReflectionProperty( $wrapped, 'modelPreferenceKeys' ) )->getValue( $wrapped ); if ( 0.2 !== $config->getTemperature() || ! is_string( $config->getSystemInstruction() ) || '' === trim( $config->getSystemInstruction() ) || array( 'model::gpt-4o-mini', 'model::gpt-4o' ) !== $preferences ) { return false; } add_filter( 'wp_ai_client_prevent_prompt', '__return_true' ); $result = $builder->generate_text(); remove_filter( 'wp_ai_client_prevent_prompt', '__return_true' ); return is_wp_error( $result ) && 'prompt_prevented' === $result->get_error_code();",
             "description": "The helper returns a usable builder without calling a provider",
             "weight": 1
           }
@@ -113,10 +112,15 @@
         "source_refs": [
           "src/wp-includes/ai-client.php",
           "src/wp-includes/ai-client/class-wp-ai-client-prompt-builder.php",
-          "src/wp-includes/php-ai-client/src/Builders/PromptBuilder.php"
+          "src/wp-includes/php-ai-client/src/Builders/PromptBuilder.php",
+          "src/wp-includes/php-ai-client/src/Providers/Models/DTO/ModelConfig.php"
         ],
         "release_focus": "7.0"
-      }
+      },
+      "exploit_solutions": [
+        "function wpbp_ai_client_002(): WP_AI_Client_Prompt_Builder { return wp_ai_client_prompt( 'Summarize WordPress.' ); }",
+        "function wpbp_ai_client_002(): WP_AI_Client_Prompt_Builder { return wp_ai_client_prompt( 'Summarize WordPress.' )->using_system_instruction( 'Be concise.' )->using_temperature( 0.2 )->using_model_preference( 'gpt-4o', 'gpt-4o-mini' ); }"
+      ]
     },
     {
       "id": "e-ai-client-003",
@@ -176,7 +180,7 @@
       "id": "e-ai-client-004",
       "category": "ai-client",
       "prompt": "Implement a function that returns the AI Client function-call name for a supplied ability name and the ability name obtained by converting it back.",
-      "expected_behavior": "`wpbp_ai_client_004()` uses `WP_AI_Client_Ability_Function_Resolver` conversion helpers to return `array( $function_name, $round_trip_ability_name )` for any supplied ability name. Runtime validation checks multiple ability names so hard-coded output does not pass.",
+      "expected_behavior": "The helper returns both the converted function name and the round-tripped ability name for multiple inputs. Numeric and associative array keys are both accepted because the prompt specifies values, not array keys.",
       "requirements": [
         "Use WP_AI_Client_Ability_Function_Resolver",
         "Use both conversion directions",
@@ -206,7 +210,7 @@
         "assertions": [
           {
             "type": "custom_assertion",
-            "code": "if ( ! function_exists( 'wpbp_ai_client_004' ) ) { return false; } $first = wpbp_ai_client_004( 'wpbp/do-thing' ); $second = wpbp_ai_client_004( 'my-plugin/export_data' ); return is_array( $first ) && is_array( $second ) && 'wpab__wpbp__do-thing' === ( $first[0] ?? null ) && 'wpbp/do-thing' === ( $first[1] ?? null ) && 'wpab__my-plugin__export_data' === ( $second[0] ?? null ) && 'my-plugin/export_data' === ( $second[1] ?? null );",
+            "code": "if ( ! function_exists( 'wpbp_ai_client_004' ) ) { return false; } $first = wpbp_ai_client_004( 'wpbp/do-thing' ); $second = wpbp_ai_client_004( 'my-plugin/export_data' ); return is_array( $first ) && is_array( $second ) && in_array( 'wpab__wpbp__do-thing', $first, true ) && in_array( 'wpbp/do-thing', $first, true ) && in_array( 'wpab__my-plugin__export_data', $second, true ) && in_array( 'my-plugin/export_data', $second, true );",
             "description": "Ability names round-trip through AI function-call names",
             "weight": 1
           }
diff --git a/datasets/suites/wp-core-v1/execution/auth.json b/datasets/suites/wp-core/execution/auth.json
similarity index 99%
rename from datasets/suites/wp-core-v1/execution/auth.json
rename to datasets/suites/wp-core/execution/auth.json
index 07ee515..d53cbaa 100644
--- a/datasets/suites/wp-core-v1/execution/auth.json
+++ b/datasets/suites/wp-core/execution/auth.json
@@ -1,6 +1,5 @@
 {
-  "id": "wp-core-execution-v1-auth",
-  "version": "1.0.1",
+  "id": "wp-core",
   "metadata": {
     "name": "WordPress Core Execution Tests - Authentication",
     "description": "Code generation tasks for WordPress authentication, password and session APIs",
diff --git a/datasets/suites/wp-core-v1/execution/block-patterns-styles.json b/datasets/suites/wp-core/execution/block-patterns-styles.json
similarity index 99%
rename from datasets/suites/wp-core-v1/execution/block-patterns-styles.json
rename to datasets/suites/wp-core/execution/block-patterns-styles.json
index 5ec25e3..6f52299 100644
--- a/datasets/suites/wp-core-v1/execution/block-patterns-styles.json
+++ b/datasets/suites/wp-core/execution/block-patterns-styles.json
@@ -1,6 +1,5 @@
 {
-  "id": "wp-core-execution-v1-block_patterns_styles",
-  "version": "1.0.1",
+  "id": "wp-core",
   "metadata": {
     "name": "WordPress Core Execution Tests - Block Patterns and Styles",
     "description": "Code generation tasks for WordPress block pattern, block style and block template registration APIs",
diff --git a/datasets/suites/wp-core-v1/execution/caching.json b/datasets/suites/wp-core/execution/caching.json
similarity index 99%
rename from datasets/suites/wp-core-v1/execution/caching.json
rename to datasets/suites/wp-core/execution/caching.json
index 85ee08e..661ecaa 100644
--- a/datasets/suites/wp-core-v1/execution/caching.json
+++ b/datasets/suites/wp-core/execution/caching.json
@@ -1,6 +1,5 @@
 {
-  "id": "wp-core-execution-v1-caching",
-  "version": "2.3.0",
+  "id": "wp-core",
   "metadata": {
     "name": "WordPress Core Execution Tests - Caching",
     "description": "Code generation tasks for WordPress Caching APIs",
diff --git a/datasets/suites/wp-core-v1/execution/comments-users.json b/datasets/suites/wp-core/execution/comments-users.json
similarity index 99%
rename from datasets/suites/wp-core-v1/execution/comments-users.json
rename to datasets/suites/wp-core/execution/comments-users.json
index de900da..4ae2055 100644
--- a/datasets/suites/wp-core-v1/execution/comments-users.json
+++ b/datasets/suites/wp-core/execution/comments-users.json
@@ -1,6 +1,5 @@
 {
-  "id": "wp-core-execution-v1-comments_users",
-  "version": "2.4.1",
+  "id": "wp-core",
   "metadata": {
     "name": "WordPress Core Execution Tests - Comments and Users",
     "description": "Code generation tasks for WordPress Comments and Users APIs",
diff --git a/datasets/suites/wp-core-v1/execution/connectors-api.json b/datasets/suites/wp-core/execution/connectors-api.json
similarity index 98%
rename from datasets/suites/wp-core-v1/execution/connectors-api.json
rename to datasets/suites/wp-core/execution/connectors-api.json
index 3544ef0..2b2001d 100644
--- a/datasets/suites/wp-core-v1/execution/connectors-api.json
+++ b/datasets/suites/wp-core/execution/connectors-api.json
@@ -1,6 +1,5 @@
 {
-  "id": "wp-core-execution-v1-connectors_api",
-  "version": "2.4.0",
+  "id": "wp-core",
   "metadata": {
     "name": "WordPress Core Execution Tests - Connectors API",
     "description": "Code generation tasks for WordPress Connectors API APIs",
@@ -183,7 +182,7 @@
       "id": "e-connectors-api-004",
       "category": "connectors-api",
       "prompt": "Register a connector with the ID 'wpbp_crm' through the Connectors API extension action: name 'WPBP CRM', type 'crm', API-key authentication with no explicit setting name.",
-      "expected_behavior": "The connector registers from a `wp_connectors_init` callback with `authentication => array( 'method' => 'api_key' )`. The runtime re-fires the action and verifies the stored connector, including the registry's auto-generated setting name `connectors_crm_wpbp_crm_api_key` and the default `plugin.is_active` callback.",
+      "expected_behavior": "The connector has the requested ID, name, type, and authentication method, and receives the registry-generated setting name. An incidental plugin callback representation is not constrained by the prompt.",
       "requirements": [
         "Register from the wp_connectors_init action",
         "Use api_key authentication without a setting name"
@@ -211,7 +210,7 @@
         "assertions": [
           {
             "type": "custom_assertion",
-            "code": "$registry = WP_Connector_Registry::get_instance(); if ( ! $registry instanceof WP_Connector_Registry ) { return false; } do_action( 'wp_connectors_init', $registry ); if ( ! wp_is_connector_registered( 'wpbp_crm' ) ) { return false; } $connector = wp_get_connector( 'wpbp_crm' ); return 'WPBP CRM' === $connector['name'] && 'crm' === $connector['type'] && 'api_key' === $connector['authentication']['method'] && 'connectors_crm_wpbp_crm_api_key' === $connector['authentication']['setting_name'] && '__return_true' === $connector['plugin']['is_active'];",
+            "code": "$registry = WP_Connector_Registry::get_instance(); if ( ! $registry instanceof WP_Connector_Registry ) { return false; } do_action( 'wp_connectors_init', $registry ); if ( ! wp_is_connector_registered( 'wpbp_crm' ) ) { return false; } $connector = wp_get_connector( 'wpbp_crm' ); return 'WPBP CRM' === $connector['name'] && 'crm' === $connector['type'] && 'api_key' === $connector['authentication']['method'] && 'connectors_crm_wpbp_crm_api_key' === $connector['authentication']['setting_name'];",
             "description": "The connector registers with the auto-generated setting name and default plugin callback",
             "weight": 1
           }
diff --git a/datasets/suites/wp-core-v1/execution/cron.json b/datasets/suites/wp-core/execution/cron.json
similarity index 99%
rename from datasets/suites/wp-core-v1/execution/cron.json
rename to datasets/suites/wp-core/execution/cron.json
index 9b24e66..8881f7e 100644
--- a/datasets/suites/wp-core-v1/execution/cron.json
+++ b/datasets/suites/wp-core/execution/cron.json
@@ -1,6 +1,5 @@
 {
-  "id": "wp-core-execution-v1-cron",
-  "version": "2.3.0",
+  "id": "wp-core",
   "metadata": {
     "name": "WordPress Core Execution Tests - Cron",
     "description": "Code generation tasks for WordPress Cron APIs",
diff --git a/datasets/suites/wp-core-v1/execution/database.json b/datasets/suites/wp-core/execution/database.json
similarity index 99%
rename from datasets/suites/wp-core-v1/execution/database.json
rename to datasets/suites/wp-core/execution/database.json
index 176d4f7..e0f8112 100644
--- a/datasets/suites/wp-core-v1/execution/database.json
+++ b/datasets/suites/wp-core/execution/database.json
@@ -1,6 +1,5 @@
 {
-  "id": "wp-core-execution-v1-database",
-  "version": "2.3.0",
+  "id": "wp-core",
   "metadata": {
     "name": "WordPress Core Execution Tests - Database",
     "description": "Code generation tasks for WordPress Database APIs",
diff --git a/datasets/suites/wp-core-v1/execution/dates-time.json b/datasets/suites/wp-core/execution/dates-time.json
similarity index 99%
rename from datasets/suites/wp-core-v1/execution/dates-time.json
rename to datasets/suites/wp-core/execution/dates-time.json
index 23f81ed..4c99914 100644
--- a/datasets/suites/wp-core-v1/execution/dates-time.json
+++ b/datasets/suites/wp-core/execution/dates-time.json
@@ -1,6 +1,5 @@
 {
-  "id": "wp-core-execution-v1-dates_time",
-  "version": "1.0.0",
+  "id": "wp-core",
   "metadata": {
     "name": "WordPress Core Execution Tests - Dates and Time",
     "description": "Timezone-aware date handling: site timezone rendering, UTC round trips, relative ages and local calendar-day queries.",
diff --git a/datasets/suites/wp-core-v1/execution/gb-block-api.json b/datasets/suites/wp-core/execution/gb-block-api.json
similarity index 99%
rename from datasets/suites/wp-core-v1/execution/gb-block-api.json
rename to datasets/suites/wp-core/execution/gb-block-api.json
index f3831ca..491cc54 100644
--- a/datasets/suites/wp-core-v1/execution/gb-block-api.json
+++ b/datasets/suites/wp-core/execution/gb-block-api.json
@@ -1,6 +1,5 @@
 {
-  "id": "wp-core-execution-v1-gb_block_api",
-  "version": "2.3.1",
+  "id": "wp-core",
   "metadata": {
     "name": "WordPress Core Execution Tests - Block API",
     "description": "Code generation tasks for WordPress Block API APIs",
diff --git a/datasets/suites/wp-core-v1/execution/gb-block-bindings.json b/datasets/suites/wp-core/execution/gb-block-bindings.json
similarity index 99%
rename from datasets/suites/wp-core-v1/execution/gb-block-bindings.json
rename to datasets/suites/wp-core/execution/gb-block-bindings.json
index 9c62eb8..4fdeee7 100644
--- a/datasets/suites/wp-core-v1/execution/gb-block-bindings.json
+++ b/datasets/suites/wp-core/execution/gb-block-bindings.json
@@ -1,6 +1,5 @@
 {
-  "id": "wp-core-execution-v1-gb_block_bindings",
-  "version": "2.3.0",
+  "id": "wp-core",
   "metadata": {
     "name": "WordPress Core Execution Tests - Block Bindings",
     "description": "Code generation tasks for WordPress Block Bindings APIs",
diff --git a/datasets/suites/wp-core-v1/execution/gb-block-editor.json b/datasets/suites/wp-core/execution/gb-block-editor.json
similarity index 99%
rename from datasets/suites/wp-core-v1/execution/gb-block-editor.json
rename to datasets/suites/wp-core/execution/gb-block-editor.json
index 75aa635..86bd8aa 100644
--- a/datasets/suites/wp-core-v1/execution/gb-block-editor.json
+++ b/datasets/suites/wp-core/execution/gb-block-editor.json
@@ -1,6 +1,5 @@
 {
-  "id": "wp-core-execution-v1-gb_block_editor",
-  "version": "2.3.0",
+  "id": "wp-core",
   "metadata": {
     "name": "WordPress Core Execution Tests - Block Editor",
     "description": "Code generation tasks for WordPress Block Editor APIs",
diff --git a/datasets/suites/wp-core-v1/execution/gb-block-hooks.json b/datasets/suites/wp-core/execution/gb-block-hooks.json
similarity index 99%
rename from datasets/suites/wp-core-v1/execution/gb-block-hooks.json
rename to datasets/suites/wp-core/execution/gb-block-hooks.json
index 79bbd8d..5e30b1c 100644
--- a/datasets/suites/wp-core-v1/execution/gb-block-hooks.json
+++ b/datasets/suites/wp-core/execution/gb-block-hooks.json
@@ -1,6 +1,5 @@
 {
-  "id": "wp-core-execution-v1-gb_block_hooks",
-  "version": "2.2.0",
+  "id": "wp-core",
   "metadata": {
     "name": "WordPress Core Execution Tests - Block Hooks",
     "description": "Code generation tasks for WordPress Block Hooks APIs",
diff --git a/datasets/suites/wp-core-v1/execution/gb-block-markup.json b/datasets/suites/wp-core/execution/gb-block-markup.json
similarity index 99%
rename from datasets/suites/wp-core-v1/execution/gb-block-markup.json
rename to datasets/suites/wp-core/execution/gb-block-markup.json
index 2ff2da8..cad454c 100644
--- a/datasets/suites/wp-core-v1/execution/gb-block-markup.json
+++ b/datasets/suites/wp-core/execution/gb-block-markup.json
@@ -1,6 +1,5 @@
 {
-  "id": "wp-core-execution-v1-gb_block_markup",
-  "version": "1.2.0",
+  "id": "wp-core",
   "metadata": {
     "name": "WordPress Core Execution Tests - Block Markup Authoring",
     "description": "Code generation tasks that require authoring serialized Gutenberg block markup as stored in post_content",
diff --git a/datasets/suites/wp-core-v1/execution/gb-font-library.json b/datasets/suites/wp-core/execution/gb-font-library.json
similarity index 99%
rename from datasets/suites/wp-core-v1/execution/gb-font-library.json
rename to datasets/suites/wp-core/execution/gb-font-library.json
index 48e8f46..686c15c 100644
--- a/datasets/suites/wp-core-v1/execution/gb-font-library.json
+++ b/datasets/suites/wp-core/execution/gb-font-library.json
@@ -1,6 +1,5 @@
 {
-  "id": "wp-core-execution-v1-gb_font_library",
-  "version": "2.2.0",
+  "id": "wp-core",
   "metadata": {
     "name": "WordPress Core Execution Tests - Font Library",
     "description": "Code generation tasks for WordPress Font Library APIs",
diff --git a/datasets/suites/wp-core-v1/execution/gb-interactivity-api.json b/datasets/suites/wp-core/execution/gb-interactivity-api.json
similarity index 99%
rename from datasets/suites/wp-core-v1/execution/gb-interactivity-api.json
rename to datasets/suites/wp-core/execution/gb-interactivity-api.json
index 5196103..e4a4fa7 100644
--- a/datasets/suites/wp-core-v1/execution/gb-interactivity-api.json
+++ b/datasets/suites/wp-core/execution/gb-interactivity-api.json
@@ -1,6 +1,5 @@
 {
-  "id": "wp-core-execution-v1-gb_interactivity_api",
-  "version": "2.3.1",
+  "id": "wp-core",
   "metadata": {
     "name": "WordPress Core Execution Tests - Interactivity API",
     "description": "Code generation tasks for WordPress Interactivity API APIs",
diff --git a/datasets/suites/wp-core-v1/execution/gb-templates-navigation.json b/datasets/suites/wp-core/execution/gb-templates-navigation.json
similarity index 99%
rename from datasets/suites/wp-core-v1/execution/gb-templates-navigation.json
rename to datasets/suites/wp-core/execution/gb-templates-navigation.json
index 0de679b..6592f6a 100644
--- a/datasets/suites/wp-core-v1/execution/gb-templates-navigation.json
+++ b/datasets/suites/wp-core/execution/gb-templates-navigation.json
@@ -1,6 +1,5 @@
 {
-  "id": "wp-core-execution-v1-gb_templates_navigation",
-  "version": "2.2.0",
+  "id": "wp-core",
   "metadata": {
     "name": "WordPress Core Execution Tests - Templates and Navigation",
     "description": "Code generation tasks for WordPress Templates and Navigation APIs",
diff --git a/datasets/suites/wp-core-v1/execution/hooks.json b/datasets/suites/wp-core/execution/hooks.json
similarity index 99%
rename from datasets/suites/wp-core-v1/execution/hooks.json
rename to datasets/suites/wp-core/execution/hooks.json
index 83b577b..4a14e84 100644
--- a/datasets/suites/wp-core-v1/execution/hooks.json
+++ b/datasets/suites/wp-core/execution/hooks.json
@@ -1,6 +1,5 @@
 {
-  "id": "wp-core-execution-v1-hooks",
-  "version": "2.3.0",
+  "id": "wp-core",
   "metadata": {
     "name": "WordPress Core Execution Tests - Hooks",
     "description": "Code generation tasks for WordPress Hooks APIs",
diff --git a/datasets/suites/wp-core-v1/execution/html-api.json b/datasets/suites/wp-core/execution/html-api.json
similarity index 99%
rename from datasets/suites/wp-core-v1/execution/html-api.json
rename to datasets/suites/wp-core/execution/html-api.json
index 133b683..f062f4e 100644
--- a/datasets/suites/wp-core-v1/execution/html-api.json
+++ b/datasets/suites/wp-core/execution/html-api.json
@@ -1,6 +1,5 @@
 {
-  "id": "wp-core-execution-v1-html_api",
-  "version": "1.0.1",
+  "id": "wp-core",
   "metadata": {
     "name": "WordPress Core Execution Tests - HTML API",
     "description": "Code generation tasks for the WordPress HTML API (Tag Processor, HTML Processor, and decoder)",
diff --git a/datasets/suites/wp-core-v1/execution/http-api.json b/datasets/suites/wp-core/execution/http-api.json
similarity index 99%
rename from datasets/suites/wp-core-v1/execution/http-api.json
rename to datasets/suites/wp-core/execution/http-api.json
index 2ac1ad0..2c667d8 100644
--- a/datasets/suites/wp-core-v1/execution/http-api.json
+++ b/datasets/suites/wp-core/execution/http-api.json
@@ -1,6 +1,5 @@
 {
-  "id": "wp-core-execution-v1-http_api",
-  "version": "1.0.0",
+  "id": "wp-core",
   "metadata": {
     "name": "WordPress Core Execution Tests - HTTP API",
     "description": "Code generation tasks for the WordPress HTTP API",
diff --git a/datasets/suites/wp-core-v1/execution/internationalization.json b/datasets/suites/wp-core/execution/internationalization.json
similarity index 99%
rename from datasets/suites/wp-core-v1/execution/internationalization.json
rename to datasets/suites/wp-core/execution/internationalization.json
index 3076f8d..9992391 100644
--- a/datasets/suites/wp-core-v1/execution/internationalization.json
+++ b/datasets/suites/wp-core/execution/internationalization.json
@@ -1,6 +1,5 @@
 {
-  "id": "wp-core-execution-v1-internationalization",
-  "version": "2.3.0",
+  "id": "wp-core",
   "metadata": {
     "name": "WordPress Core Execution Tests - Internationalization",
     "description": "Code generation tasks for WordPress Internationalization APIs",
diff --git a/datasets/suites/wp-core-v1/execution/mail-url.json b/datasets/suites/wp-core/execution/mail-url.json
similarity index 99%
rename from datasets/suites/wp-core-v1/execution/mail-url.json
rename to datasets/suites/wp-core/execution/mail-url.json
index 082cf7c..42aa5c0 100644
--- a/datasets/suites/wp-core-v1/execution/mail-url.json
+++ b/datasets/suites/wp-core/execution/mail-url.json
@@ -1,6 +1,5 @@
 {
-  "id": "wp-core-execution-v1-mail_url",
-  "version": "2.3.0",
+  "id": "wp-core",
   "metadata": {
     "name": "WordPress Core Execution Tests - Mail and URL Handling",
     "description": "Code generation tasks for WordPress Mail and URL Handling APIs",
diff --git a/datasets/suites/wp-core-v1/execution/media.json b/datasets/suites/wp-core/execution/media.json
similarity index 99%
rename from datasets/suites/wp-core-v1/execution/media.json
rename to datasets/suites/wp-core/execution/media.json
index 64001df..72eacfd 100644
--- a/datasets/suites/wp-core-v1/execution/media.json
+++ b/datasets/suites/wp-core/execution/media.json
@@ -1,6 +1,5 @@
 {
-  "id": "wp-core-execution-v1-media",
-  "version": "2.3.0",
+  "id": "wp-core",
   "metadata": {
     "name": "WordPress Core Execution Tests - Media",
     "description": "Code generation tasks for WordPress Media APIs",
diff --git a/datasets/suites/wp-core-v1/execution/performance.json b/datasets/suites/wp-core/execution/performance.json
similarity index 99%
rename from datasets/suites/wp-core-v1/execution/performance.json
rename to datasets/suites/wp-core/execution/performance.json
index 6f7abed..34aa341 100644
--- a/datasets/suites/wp-core-v1/execution/performance.json
+++ b/datasets/suites/wp-core/execution/performance.json
@@ -1,6 +1,5 @@
 {
-  "id": "wp-core-execution-v1-performance",
-  "version": "1.0.1",
+  "id": "wp-core",
   "metadata": {
     "name": "WordPress Core Execution Tests - Performance",
     "description": "Code generation tasks for WordPress asset loading, script modules, speculative loading, and media performance APIs",
diff --git a/datasets/suites/wp-core-v1/execution/plugin-lifecycle.json b/datasets/suites/wp-core/execution/plugin-lifecycle.json
similarity index 99%
rename from datasets/suites/wp-core-v1/execution/plugin-lifecycle.json
rename to datasets/suites/wp-core/execution/plugin-lifecycle.json
index 219ca87..c253a8b 100644
--- a/datasets/suites/wp-core-v1/execution/plugin-lifecycle.json
+++ b/datasets/suites/wp-core/execution/plugin-lifecycle.json
@@ -1,6 +1,5 @@
 {
-  "id": "wp-core-execution-v1-plugin_lifecycle",
-  "version": "1.0.0",
+  "id": "wp-core",
   "metadata": {
     "name": "WordPress Core Execution Tests - Plugin Lifecycle",
     "description": "Plugin artifact tasks for WordPress plugin activation, upgrade and plugin screen integration",
diff --git a/datasets/suites/wp-core-v1/execution/post-meta.json b/datasets/suites/wp-core/execution/post-meta.json
similarity index 99%
rename from datasets/suites/wp-core-v1/execution/post-meta.json
rename to datasets/suites/wp-core/execution/post-meta.json
index c13341e..2cf21f4 100644
--- a/datasets/suites/wp-core-v1/execution/post-meta.json
+++ b/datasets/suites/wp-core/execution/post-meta.json
@@ -1,6 +1,5 @@
 {
-  "id": "wp-core-execution-v1-post_meta",
-  "version": "2.2.0",
+  "id": "wp-core",
   "metadata": {
     "name": "WordPress Core Execution Tests - Post Meta",
     "description": "Code generation tasks for WordPress Post Meta APIs",
diff --git a/datasets/suites/wp-core-v1/execution/post-types-taxonomy.json b/datasets/suites/wp-core/execution/post-types-taxonomy.json
similarity index 99%
rename from datasets/suites/wp-core-v1/execution/post-types-taxonomy.json
rename to datasets/suites/wp-core/execution/post-types-taxonomy.json
index 9a19c8a..ddac1b3 100644
--- a/datasets/suites/wp-core-v1/execution/post-types-taxonomy.json
+++ b/datasets/suites/wp-core/execution/post-types-taxonomy.json
@@ -1,6 +1,5 @@
 {
-  "id": "wp-core-execution-v1-post_types_taxonomy",
-  "version": "2.4.0",
+  "id": "wp-core",
   "metadata": {
     "name": "WordPress Core Execution Tests - Post Types and Taxonomy",
     "description": "Code generation tasks for WordPress Post Types and Taxonomy APIs",
diff --git a/datasets/suites/wp-core-v1/execution/posts-lifecycle.json b/datasets/suites/wp-core/execution/posts-lifecycle.json
similarity index 99%
rename from datasets/suites/wp-core-v1/execution/posts-lifecycle.json
rename to datasets/suites/wp-core/execution/posts-lifecycle.json
index dbcb69c..a6bf0f7 100644
--- a/datasets/suites/wp-core-v1/execution/posts-lifecycle.json
+++ b/datasets/suites/wp-core/execution/posts-lifecycle.json
@@ -1,6 +1,5 @@
 {
-  "id": "wp-core-execution-v1-posts_lifecycle",
-  "version": "1.0.0",
+  "id": "wp-core",
   "metadata": {
     "name": "WordPress Core Execution Tests - Posts Lifecycle",
     "description": "Code generation tasks for the WordPress post creation, publication, revision and deletion lifecycle",
diff --git a/datasets/suites/wp-core-v1/execution/queries.json b/datasets/suites/wp-core/execution/queries.json
similarity index 99%
rename from datasets/suites/wp-core-v1/execution/queries.json
rename to datasets/suites/wp-core/execution/queries.json
index e0b9709..ea2b82f 100644
--- a/datasets/suites/wp-core-v1/execution/queries.json
+++ b/datasets/suites/wp-core/execution/queries.json
@@ -1,6 +1,5 @@
 {
-  "id": "wp-core-execution-v1-queries",
-  "version": "2.3.0",
+  "id": "wp-core",
   "metadata": {
     "name": "WordPress Core Execution Tests - Queries",
     "description": "Code generation tasks for WordPress Queries APIs",
diff --git a/datasets/suites/wp-core-v1/execution/rest-api.json b/datasets/suites/wp-core/execution/rest-api.json
similarity index 99%
rename from datasets/suites/wp-core-v1/execution/rest-api.json
rename to datasets/suites/wp-core/execution/rest-api.json
index bc0c868..84f54bf 100644
--- a/datasets/suites/wp-core-v1/execution/rest-api.json
+++ b/datasets/suites/wp-core/execution/rest-api.json
@@ -1,6 +1,5 @@
 {
-  "id": "wp-core-execution-v1-rest_api",
-  "version": "2.3.0",
+  "id": "wp-core",
   "metadata": {
     "name": "WordPress Core Execution Tests - REST API",
     "description": "Code generation tasks for WordPress REST API APIs",
diff --git a/datasets/suites/wp-core-v1/execution/rewrite-permalinks.json b/datasets/suites/wp-core/execution/rewrite-permalinks.json
similarity index 99%
rename from datasets/suites/wp-core-v1/execution/rewrite-permalinks.json
rename to datasets/suites/wp-core/execution/rewrite-permalinks.json
index b91d03e..fd14923 100644
--- a/datasets/suites/wp-core-v1/execution/rewrite-permalinks.json
+++ b/datasets/suites/wp-core/execution/rewrite-permalinks.json
@@ -1,6 +1,5 @@
 {
-  "id": "wp-core-execution-v1-rewrite_permalinks",
-  "version": "2.3.0",
+  "id": "wp-core",
   "metadata": {
     "name": "WordPress Core Execution Tests - Rewrite and Permalinks",
     "description": "Code generation tasks for WordPress Rewrite and Permalinks APIs",
diff --git a/datasets/suites/wp-core-v1/execution/roles-caps.json b/datasets/suites/wp-core/execution/roles-caps.json
similarity index 99%
rename from datasets/suites/wp-core-v1/execution/roles-caps.json
rename to datasets/suites/wp-core/execution/roles-caps.json
index 3b3d5b7..63057eb 100644
--- a/datasets/suites/wp-core-v1/execution/roles-caps.json
+++ b/datasets/suites/wp-core/execution/roles-caps.json
@@ -1,6 +1,5 @@
 {
-  "id": "wp-core-execution-v1-roles_caps",
-  "version": "2.4.0",
+  "id": "wp-core",
   "metadata": {
     "name": "WordPress Core Execution Tests - Roles and Capabilities",
     "description": "Code generation tasks for WordPress Roles and Capabilities APIs",
diff --git a/datasets/suites/wp-core-v1/execution/scripts-styles.json b/datasets/suites/wp-core/execution/scripts-styles.json
similarity index 99%
rename from datasets/suites/wp-core-v1/execution/scripts-styles.json
rename to datasets/suites/wp-core/execution/scripts-styles.json
index 0c80c36..f103a27 100644
--- a/datasets/suites/wp-core-v1/execution/scripts-styles.json
+++ b/datasets/suites/wp-core/execution/scripts-styles.json
@@ -1,6 +1,5 @@
 {
-  "id": "wp-core-execution-v1-scripts_styles",
-  "version": "2.2.0",
+  "id": "wp-core",
   "metadata": {
     "name": "WordPress Core Execution Tests - Scripts and Styles",
     "description": "Code generation tasks for WordPress Scripts and Styles APIs",
diff --git a/datasets/suites/wp-core-v1/execution/security.json b/datasets/suites/wp-core/execution/security.json
similarity index 99%
rename from datasets/suites/wp-core-v1/execution/security.json
rename to datasets/suites/wp-core/execution/security.json
index c62e499..dcaa037 100644
--- a/datasets/suites/wp-core-v1/execution/security.json
+++ b/datasets/suites/wp-core/execution/security.json
@@ -1,6 +1,5 @@
 {
-  "id": "wp-core-execution-v1-security",
-  "version": "2.3.0",
+  "id": "wp-core",
   "metadata": {
     "name": "WordPress Core Execution Tests - Security",
     "description": "Code generation tasks for WordPress Security APIs",
diff --git a/datasets/suites/wp-core-v1/execution/settings-options.json b/datasets/suites/wp-core/execution/settings-options.json
similarity index 99%
rename from datasets/suites/wp-core-v1/execution/settings-options.json
rename to datasets/suites/wp-core/execution/settings-options.json
index 03a0e6d..e14c198 100644
--- a/datasets/suites/wp-core-v1/execution/settings-options.json
+++ b/datasets/suites/wp-core/execution/settings-options.json
@@ -1,6 +1,5 @@
 {
-  "id": "wp-core-execution-v1-settings_options",
-  "version": "2.3.0",
+  "id": "wp-core",
   "metadata": {
     "name": "WordPress Core Execution Tests - Settings and Options",
     "description": "Code generation tasks for WordPress Settings and Options APIs",
diff --git a/datasets/suites/wp-core-v1/execution/shortcodes.json b/datasets/suites/wp-core/execution/shortcodes.json
similarity index 99%
rename from datasets/suites/wp-core-v1/execution/shortcodes.json
rename to datasets/suites/wp-core/execution/shortcodes.json
index f05f35e..7584476 100644
--- a/datasets/suites/wp-core-v1/execution/shortcodes.json
+++ b/datasets/suites/wp-core/execution/shortcodes.json
@@ -1,6 +1,5 @@
 {
-  "id": "wp-core-execution-v1-shortcodes",
-  "version": "2.2.0",
+  "id": "wp-core",
   "metadata": {
     "name": "WordPress Core Execution Tests - Shortcodes",
     "description": "Code generation tasks for WordPress Shortcodes APIs",
diff --git a/datasets/suites/wp-core-v1/execution/theme-templates.json b/datasets/suites/wp-core/execution/theme-templates.json
similarity index 99%
rename from datasets/suites/wp-core-v1/execution/theme-templates.json
rename to datasets/suites/wp-core/execution/theme-templates.json
index 67defdc..d208f60 100644
--- a/datasets/suites/wp-core-v1/execution/theme-templates.json
+++ b/datasets/suites/wp-core/execution/theme-templates.json
@@ -1,6 +1,5 @@
 {
-  "id": "wp-core-execution-v1-theme_templates",
-  "version": "1.0.1",
+  "id": "wp-core",
   "metadata": {
     "name": "WordPress Core Execution Tests - Theme Templates",
     "description": "Code generation tasks for WordPress theme, block template, global styles, and template rendering APIs",
diff --git a/python/tests/test_artifacts.py b/python/tests/test_artifacts.py
index 24df833..53cfdc5 100644
--- a/python/tests/test_artifacts.py
+++ b/python/tests/test_artifacts.py
@@ -39,7 +39,7 @@
 def _plugin_test(test_id: str = "e-plugin-001") -> ExecutionTest:
     return ExecutionTest(
         id=test_id,
-        suite="wp-core-v1",
+        suite="wp-core",
         prompt="Build a plugin.",
         expected_behavior="expected",
         category="plugins",
@@ -143,7 +143,7 @@ def test_prompt_instructions_differ_by_kind() -> None:
 
 def _config(tmp_path: Path, **run_overrides: object) -> HarnessConfig:
     return HarnessConfig(
-        dataset=DatasetConfig(source="local", name="wp-core-v1"),
+        dataset=DatasetConfig(source="local", name="wp-core"),
         model=ModelConfig(name="test-model"),
         grader=GraderConfig(kind="cli"),
         run=RunConfig.model_validate(dict(run_overrides)),
diff --git a/python/tests/test_benchmark_releases.py b/python/tests/test_benchmark_releases.py
new file mode 100644
index 0000000..5a7cdea
--- /dev/null
+++ b/python/tests/test_benchmark_releases.py
@@ -0,0 +1,347 @@
+"""Release identity, immutable history, and safe website ingestion."""
+from __future__ import annotations
+
+import copy
+import json
+import subprocess
+from pathlib import Path
+from typing import Any
+
+import pytest
+from typer.testing import CliRunner
+
+from wp_bench.benchmark import (
+    HARNESS_INPUTS,
+    capture_identity,
+    digest,
+    evaluation_snapshot,
+    freeze_release,
+    load_release,
+    publication_issues,
+    read_json,
+)
+from wp_bench.benchmark_cli import app, check_releases, export_results
+from wp_bench.config import (
+    BenchmarkConfig,
+    DatasetConfig,
+    HarnessConfig,
+    ModelConfig,
+    RunConfig,
+)
+from wp_bench.datasets import _parse_execution_suite
+
+
+@pytest.fixture
+def release_root(tmp_path: Path) -> Path:
+    root = tmp_path / "repo"
+    definition = {
+        "schema_version": "1.0", "wordpress_version": "7.1", "scoring_version": "3.0",
+        "suites": ["wp-core", "wp-cli"],
+        "categories": [{"id": "core", "label": "Core"}, {"id": "wp-cli", "label": "WP-CLI"}],
+        "category_mapping": {"wp-core": {"hooks": "core"}, "wp-cli": {"cli": "wp-cli"}},
+        "protocol": {"model": {"temperature": 1.0, "top_p": None, "max_tokens": 16384},
+                     "grader": {"kind": "docker", "timeout_seconds": 90}, "repeats": 2},
+        "runtime_image_ids": ["sha256:approved"],
+    }
+    (root / "benchmark").mkdir(parents=True)
+    (root / "benchmark/definition.json").write_text(json.dumps(definition))
+    for suite, category in [("wp-core", "hooks"), ("wp-cli", "cli")]:
+        directory = root / "datasets/suites" / suite / "execution"
+        directory.mkdir(parents=True)
+        (directory / "tasks.json").write_text(json.dumps({"id": suite, "tests": [{
+            "id": f"e-{category}-001", "prompt": "Implement the requested WordPress behavior.",
+            "requirements": [], "category": category, "runtime_checks": {"assertions": []},
+        }]}))
+    (root / "runtime").mkdir()
+    (root / "runtime/verify-runtime.php").write_text("<?php // trusted grader\n")
+    (root / "python/wp_bench").mkdir(parents=True)
+    for name in HARNESS_INPUTS:
+        (root / "python/wp_bench" / name).write_text("# fixture\n")
+    freeze_release("v1", root)
+    return root
+
+
+def config(suite: str = "wp-core", pinned: bool = False) -> HarnessConfig:
+    return HarnessConfig(
+        benchmark=BenchmarkConfig(version="v1" if pinned else None),
+        dataset=DatasetConfig(source="local", name=suite),
+        model=ModelConfig(name="model-a", temperature=1, max_tokens=16384),
+        run=RunConfig(suite=suite),
+    )
+
+
+def result(root: Path, suite: str, run_id: str, name: str = "model-a") -> dict[str, Any]:
+    cfg = config(suite)
+    tests = _parse_execution_suite(root / "datasets/suites" / suite / "execution/tasks.json")
+    identity = capture_identity(cfg, tests, root=root)
+    identity.bind_runtime("sha256:approved")
+    records = [{"test_id": test.id, "category": test.category, "mode": "model",
+                "model": {"name": name, "temperature": 1.0, "top_p": None, "max_tokens": 16384},
+                "variant": {"kind": "none"}, "scores": {"execution_pass": True},
+                "error": None, "run_id": run_id,
+                "benchmark": {**identity.record_info(), "task_sha256": identity.tasks[test.id]}}
+               for test in tests]
+    return {"metadata": {"suite": suite, "run_id": run_id, "benchmark": identity.metadata(records),
+                         "grader": cfg.grader.model_dump(mode="json"), "scoring_version": "3.0",
+                         "runtime_isolation": "reset_per_test", "runtime_image_id": "sha256:approved"},
+            "results": records}
+
+
+def inputs(root: Path, tmp_path: Path, name: str = "model-a") -> list[Path]:
+    paths = []
+    for suite in ["wp-core", "wp-cli"]:
+        for repeat in [1, 2]:
+            path = tmp_path / f"{name}-{suite}-{repeat}.json"
+            path.write_text(json.dumps(result(root, suite, f"{name}-{suite}-{repeat}", name)))
+            paths.append(path)
+    return paths
+
+
+def test_one_version_freezes_every_dataset_and_the_runtime(release_root: Path) -> None:
+    manifest = load_release(root=release_root)
+    assert manifest["version"] == "v1"
+    assert set(manifest["evaluation"]["components"]) == {"wp-core", "wp-cli"}
+    assert manifest["fingerprint"] == digest(evaluation_snapshot(release_root))
+    assert check_releases(release_root)["working_tree"] == "released"
+
+
+@pytest.mark.parametrize("relative", [
+    "datasets/suites/wp-core/execution/tasks.json", "datasets/suites/wp-cli/execution/tasks.json",
+    "runtime/verify-runtime.php", "python/wp_bench/scoring.py", "benchmark/definition.json",
+])
+def test_any_component_change_requires_a_new_release(release_root: Path, relative: str) -> None:
+    path = release_root / relative
+    if relative == "benchmark/definition.json":
+        definition = read_json(path)
+        definition["protocol"]["grader"]["timeout_seconds"] = 120
+        path.write_text(json.dumps(definition))
+    else:
+        path.write_text(path.read_text() + "\n")
+    # JSON formatting is conservatively part of a frozen source snapshot too.
+    tests = _parse_execution_suite(release_root / "datasets/suites/wp-core/execution/tasks.json")
+    identity = capture_identity(config(), tests, root=release_root)
+    assert identity.record_info()["version"] is None
+    assert identity.record_info()["status"] == "unreleased"
+    with pytest.raises(ValueError, match="Cannot run as v1"):
+        capture_identity(config(pinned=True), tests, root=release_root)
+    old = load_release(root=release_root)
+    new = freeze_release("v2", release_root)
+    assert old["fingerprint"] != new["fingerprint"]
+    assert load_release("v1", release_root) == old
+
+
+def test_model_additions_do_not_create_a_benchmark_version(release_root: Path) -> None:
+    before = load_release(root=release_root)
+    cfg = config()
+    cfg.model = ModelConfig(name="another-model", temperature=1, max_tokens=16384)
+    tests = _parse_execution_suite(release_root / "datasets/suites/wp-core/execution/tasks.json")
+    assert capture_identity(cfg, tests, root=release_root).record_info()["version"] == "v1"
+    assert load_release(root=release_root) == before
+    with pytest.raises(ValueError, match="unchanged"):
+        freeze_release("v2", release_root)
+
+
+def test_changed_loaded_hub_task_cannot_claim_a_release(release_root: Path) -> None:
+    tests = _parse_execution_suite(release_root / "datasets/suites/wp-core/execution/tasks.json")
+    tests[0].prompt = "A different prompt, from an unpinned Hub revision"
+    with pytest.raises(ValueError, match="Loaded tasks differ"):
+        capture_identity(config(pinned=True), tests, root=release_root)
+
+
+def test_model_protocol_change_is_not_v1(release_root: Path) -> None:
+    cfg = config(pinned=True)
+    assert cfg.model is not None
+    cfg.model.temperature = 0
+    tests = _parse_execution_suite(release_root / "datasets/suites/wp-core/execution/tasks.json")
+    with pytest.raises(ValueError, match="Model settings"):
+        capture_identity(cfg, tests, root=release_root)
+
+
+def test_runtime_is_verified_before_pinned_generation(release_root: Path) -> None:
+    tests = _parse_execution_suite(release_root / "datasets/suites/wp-core/execution/tasks.json")
+    identity = capture_identity(config(pinned=True), tests, root=release_root)
+    with pytest.raises(ValueError, match="Runtime image"):
+        identity.bind_runtime("sha256:unapproved", required=True)
+
+
+def test_frozen_releases_cannot_be_overwritten_or_relabelled(release_root: Path) -> None:
+    path = release_root / "benchmark/releases/v1.json"
+    original = path.read_bytes()
+    with pytest.raises(ValueError, match="immutable"):
+        freeze_release("v1", release_root)
+    with pytest.raises(ValueError, match="Next benchmark release must be v2"):
+        freeze_release("v3", release_root)
+    assert path.read_bytes() == original
+    changed = json.loads(original)
+    changed["evaluation"]["definition"]["wordpress_version"] = "another-version"
+    path.write_text(json.dumps(changed))
+    with pytest.raises(ValueError, match="Invalid benchmark release"):
+        load_release(root=release_root)
+
+
+@pytest.mark.parametrize("change", ["partial", "duplicate", "error", "reference", "skills", "mixed-version",
+                                    "wrong-image", "wrong-grader", "wrong-task", "wrong-run", "ungraded"])
+def test_unsafe_results_cannot_enter_a_leaderboard(release_root: Path, change: str) -> None:
+    payload = result(release_root, "wp-core", "run-one")
+    assert publication_issues(payload, load_release(root=release_root)) == []
+    record = payload["results"][0]
+    if change == "partial":
+        payload["results"] = []
+    elif change == "duplicate":
+        payload["results"].append(copy.deepcopy(record))
+    elif change == "error":
+        record["error"] = {"type": "provider_error"}
+    elif change == "reference":
+        record["mode"] = "reference_solution"
+    elif change == "skills":
+        record["variant"]["kind"] = "skills"
+    elif change == "mixed-version":
+        record["benchmark"]["version"] = "v2"
+    elif change == "wrong-image":
+        payload["metadata"]["runtime_image_id"] = "sha256:another"
+    elif change == "wrong-grader":
+        payload["metadata"]["grader"]["timeout_seconds"] = 120
+    elif change == "wrong-task":
+        record["benchmark"]["task_sha256"] = "different"
+    elif change == "wrong-run":
+        record["run_id"] = "another-run"
+    elif change == "ungraded":
+        record["scores"]["execution_pass"] = None
+    assert publication_issues(payload, load_release(root=release_root))
+
+
+def test_export_preserves_old_snapshots_and_is_idempotent(release_root: Path, tmp_path: Path) -> None:
+    paths = inputs(release_root, tmp_path)
+    output = tmp_path / "public"
+    first = export_results(paths, output, root=release_root)
+    original = first.read_bytes()
+    assert export_results(paths, output, root=release_root) == first
+    assert first.read_bytes() == original
+    bundle = json.loads(original)
+    assert bundle["benchmark"]["version"] == "v1"
+    assert len(bundle["models"][0]["answers"]) == 4
+    second = export_results(paths + inputs(release_root, tmp_path, "model-b"), output, root=release_root)
+    assert second != first
+    assert first.read_bytes() == original
+    catalog = read_json(output / "results.json")
+    assert len(catalog["benchmarks"]["v1"]["snapshots"]) == 2
+
+
+def test_export_requires_both_complete_suites_and_independent_repeats(release_root: Path, tmp_path: Path) -> None:
+    paths = inputs(release_root, tmp_path)
+    with pytest.raises(ValueError, match="independent complete runs"):
+        export_results(paths[:-1], tmp_path / "public", root=release_root)
+    with pytest.raises(ValueError, match="Duplicate run identity"):
+        export_results(paths + [paths[0]], tmp_path / "public", root=release_root)
+    assert not (tmp_path / "public").exists()
+
+
+def test_export_supports_multi_model_result_envelopes(release_root: Path, tmp_path: Path) -> None:
+    paths = inputs(release_root, tmp_path)
+    for path in paths:
+        payload = read_json(path)
+        metadata = payload["metadata"]
+        wrapped = {"metadata": metadata, "models": {"model-a": {
+            "run_id": metadata["run_id"], "benchmark": metadata["benchmark"], "results": payload["results"],
+        }}}
+        path.write_text(json.dumps(wrapped))
+    destination = export_results(paths, tmp_path / "public", root=release_root)
+    assert len(read_json(destination)["models"]) == 1
+
+
+def test_ci_detects_changes_even_when_manifest_hashes_are_recomputed(release_root: Path) -> None:
+    def git(*args: str) -> str:
+        return subprocess.run(["git", *args], cwd=release_root, check=True,
+                              capture_output=True, text=True).stdout.strip()
+    git("init")
+    git("add", ".")
+    git("-c", "user.name=Test", "-c", "user.email=test@example.invalid", "commit", "-m", "initial")
+    base = git("rev-parse", "HEAD")
+    assert check_releases(release_root, base)["working_tree"] == "released"
+    path = release_root / "benchmark/releases/v1.json"
+    manifest = read_json(path)
+    manifest["evaluation"]["definition"]["wordpress_version"] = "changed"
+    manifest["fingerprint"] = digest(manifest["evaluation"])
+    path.write_text(json.dumps(manifest))
+    catalog_path = release_root / "benchmark/releases.json"
+    catalog = read_json(catalog_path)
+    catalog["releases"][0]["fingerprint"] = manifest["fingerprint"]
+    catalog_path.write_text(json.dumps(catalog))
+    with pytest.raises(ValueError, match="Published benchmark manifest changed"):
+        check_releases(release_root, base)
+
+
+def test_cli_exposes_json_contract_without_a_model_call(release_root: Path, monkeypatch: pytest.MonkeyPatch) -> None:
+    monkeypatch.setattr("wp_bench.benchmark_cli.load_release", lambda version=None: load_release(version, release_root))
+    response = CliRunner().invoke(app, ["show", "--version", "v1"])
+    assert response.exit_code == 0
+    assert json.loads(response.stdout)["version"] == "v1"
+
+
+
+def test_export_refuses_a_corrupted_existing_snapshot(release_root: Path, tmp_path: Path) -> None:
+    paths = inputs(release_root, tmp_path)
+    output = tmp_path / "public"
+    destination = export_results(paths, output, root=release_root)
+    snapshot = read_json(destination)
+    snapshot["models"][0]["answers"][0]["passed"] = False
+    destination.write_text(json.dumps(snapshot))
+    with pytest.raises(ValueError, match="different fingerprint"):
+        export_results(paths, output, root=release_root)
+
+
+@pytest.mark.parametrize("field", ["repeats", "images", "categories", "mapping", "temperature"])
+def test_freeze_rejects_an_invalid_recipe(release_root: Path, field: str) -> None:
+    path = release_root / "benchmark/definition.json"
+    definition = read_json(path)
+    if field == "repeats":
+        definition["protocol"]["repeats"] = 0
+    elif field == "images":
+        definition["runtime_image_ids"] = []
+    elif field == "categories":
+        definition["categories"].append(definition["categories"][0])
+    elif field == "mapping":
+        del definition["category_mapping"]["wp-cli"]
+    elif field == "temperature":
+        definition["protocol"]["model"]["temperature"] = 20
+    path.write_text(json.dumps(definition))
+    with pytest.raises(ValueError):
+        freeze_release("v2", release_root)
+    assert not (release_root / "benchmark/releases/v2.json").exists()
+
+
+def test_harness_streams_the_same_verified_identity_as_the_completed_payload(
+    release_root: Path, tmp_path: Path, monkeypatch: pytest.MonkeyPatch,
+) -> None:
+    from conftest import fake_generation
+
+    from wp_bench.config import OutputConfig
+    from wp_bench.core import BenchmarkRunner
+    from wp_bench.environment import ExecutionResult
+
+    cfg = config(pinned=True)
+    cfg.output = OutputConfig(path=tmp_path / "complete.json", jsonl_path=tmp_path / "stream.jsonl")
+    tests = _parse_execution_suite(release_root / "datasets/suites/wp-core/execution/tasks.json")
+    monkeypatch.setattr("wp_bench.core.load_tests", lambda config: tests)
+    monkeypatch.setattr("wp_bench.core.capture_identity", lambda config, tests: capture_identity(config, tests, root=release_root))
+
+    class Environment:
+        image_id = "sha256:approved"
+
+        def setup(self, **kwargs: Any) -> None: ...
+        def reset(self) -> None: ...
+        def release(self) -> None: ...
+        def close(self) -> None: ...
+        def execute_artifact(self, artifact: Any, spec: Any) -> ExecutionResult:
+            return ExecutionResult(success=True, raw={"success": True, "runtime": {"score": 1.0}, "static": {"score": 1.0}}, stdout="", stderr="")
+
+    runner = BenchmarkRunner(cfg)
+    runner.environment = Environment()  # type: ignore[assignment]
+    monkeypatch.setattr(runner.model, "generate_with_metadata", lambda prompt: fake_generation("<?php echo 1;"))
+    payload = runner.run()
+    assert payload["metadata"]["benchmark"]["version"] == "v1"
+    assert payload["metadata"]["benchmark"]["full_suite"] is True
+    streamed = [json.loads(line) for line in next(tmp_path.glob("stream_*.jsonl")).read_text().splitlines()]
+    assert streamed == payload["results"]
+    assert publication_issues(payload, load_release(root=release_root)) == []
+    assert streamed[0]["run_id"] == payload["metadata"]["run_id"]
diff --git a/python/tests/test_config_behavior.py b/python/tests/test_config_behavior.py
index 3d7b5e1..eca4916 100644
--- a/python/tests/test_config_behavior.py
+++ b/python/tests/test_config_behavior.py
@@ -56,7 +56,7 @@ def test_example_config_loads() -> None:
     """The shipped example must only use supported fields."""
     example = PROJECT_ROOT / "wp-bench.example.yaml"
     config = HarnessConfig.from_file(example)
-    assert config.run.suite == "wp-core-v1"
+    assert config.run.suite == "wp-core"
     assert config.grader.kind in ("docker", "cli")
 
 
@@ -67,7 +67,7 @@ def test_skip_runtime_controls_verification_payload() -> None:
 
     test = ExecutionTest(
         id="e-one",
-        suite="wp-core-v1",
+        suite="wp-core",
         prompt="Prompt",
         expected_behavior="expected",
         category="general",
@@ -101,7 +101,7 @@ def test_skipped_dimension_not_scored() -> None:
 
     test = ExecutionTest(
         id="e-one",
-        suite="wp-core-v1",
+        suite="wp-core",
         prompt="Prompt",
         expected_behavior="expected",
         category="general",
diff --git a/python/tests/test_continue_on_error.py b/python/tests/test_continue_on_error.py
index 105cffd..ba97527 100644
--- a/python/tests/test_continue_on_error.py
+++ b/python/tests/test_continue_on_error.py
@@ -23,7 +23,7 @@
 def _execution_test(test_id: str) -> ExecutionTest:
     return ExecutionTest(
         id=test_id,
-        suite="wp-core-v1",
+        suite="wp-core",
         prompt="Prompt",
         expected_behavior="expected",
         category="general",
@@ -47,7 +47,7 @@ def _passing_result() -> ExecutionResult:
 
 def _config(tmp_path: Path, **run_overrides: Any) -> HarnessConfig:
     return HarnessConfig(
-        dataset=DatasetConfig(source="local", name="wp-core-v1"),
+        dataset=DatasetConfig(source="local", name="wp-core"),
         model=ModelConfig(name="test-model"),
         grader=GraderConfig(kind="cli"),
         run=RunConfig(**run_overrides),
diff --git a/python/tests/test_dataset_filtering.py b/python/tests/test_dataset_filtering.py
index 488435e..7b5c713 100644
--- a/python/tests/test_dataset_filtering.py
+++ b/python/tests/test_dataset_filtering.py
@@ -13,7 +13,7 @@
 def _execution_test(test_id: str, category: str = "general") -> ExecutionTest:
     return ExecutionTest(
         id=test_id,
-        suite="wp-core-v1",
+        suite="wp-core",
         prompt="Prompt",
         expected_behavior="expected",
         category=category,
@@ -58,7 +58,7 @@ def test_filter_tests_by_ids_rejects_unknown_ids() -> None:
 def test_zero_selected_tests_fails_loudly() -> None:
     """An empty selection (missing suite, execution-less dataset) must not
     produce a vacuous successful run."""
-    config = HarnessConfig.model_validate({"dataset": {"source": "local", "name": "wp-core-v1"}})
+    config = HarnessConfig.model_validate({"dataset": {"source": "local", "name": "wp-core"}})
     with pytest.raises(ValueError, match="No execution tests selected"):
         select_run_tests([], config)
 
@@ -137,7 +137,7 @@ def test_dry_run_category_filters_selected_test_ids(monkeypatch: pytest.MonkeyPa
 def test_category_filter_rejects_mixed_known_and_unknown_categories() -> None:
     config = HarnessConfig.model_validate(
         {
-            "dataset": {"source": "local", "name": "wp-core-v1"},
+            "dataset": {"source": "local", "name": "wp-core"},
             "run": {"categories": ["hooks", "not-a-category"]},
         }
     )
diff --git a/python/tests/test_dataset_metadata.py b/python/tests/test_dataset_metadata.py
index 0199fb2..300e231 100644
--- a/python/tests/test_dataset_metadata.py
+++ b/python/tests/test_dataset_metadata.py
@@ -30,7 +30,7 @@
 
 def test_local_parser_preserves_execution_task_metadata() -> None:
     """Per-task provenance fields survive local loading at the top level."""
-    tests = load_tests(DatasetConfig(source="local", name="wp-core-v1"))
+    tests = load_tests(DatasetConfig(source="local", name="wp-core"))
 
     with_refs = [t for t in tests if t.metadata.get("source_refs")]
     assert with_refs, "expected execution tasks with source_refs metadata"
@@ -73,7 +73,7 @@ def test_huggingface_loader_preserves_metadata(monkeypatch: pytest.MonkeyPatch)
     rows = [
         {
             "id": "e-one",
-            "suite": "wp-core-v1",
+            "suite": "wp-core",
             "test_kind": "execution",
             "prompt": "Prompt",
             "requirements": "[]",
@@ -84,7 +84,7 @@ def test_huggingface_loader_preserves_metadata(monkeypatch: pytest.MonkeyPatch)
         },
         {
             "id": "k-one",
-            "suite": "wp-core-v1",
+            "suite": "wp-core",
             "test_kind": "knowledge",
             "prompt": "Prompt",
             "requirements": "[]",
@@ -114,9 +114,9 @@ def test_export_dataset_writes_metadata_column() -> None:
     finally:
         sys.path.pop(0)
 
-    rows = load_suite("wp-core-v1")
+    rows = load_suite("wp-core")
 
-    assert rows, "expected exported rows for wp-core-v1"
+    assert rows, "expected exported rows for wp-core"
     assert all("metadata" in row for row in rows)
     execution_rows = [r for r in rows if r["test_kind"] == "execution"]
     assert any('"source_refs"' in r["metadata"] for r in execution_rows)
@@ -129,7 +129,7 @@ def test_result_record_includes_task_metadata(
     """Task provenance is auditable from every result record."""
     test = ExecutionTest(
         id="e-one",
-        suite="wp-core-v1",
+        suite="wp-core",
         prompt="Prompt",
         expected_behavior="expected",
         category="general",
@@ -141,7 +141,7 @@ def test_result_record_includes_task_metadata(
         metadata={"source_refs": ["plugin.php"], "release_focus": "6.9", "suite_metadata": {}},
     )
     config = HarnessConfig(
-        dataset=DatasetConfig(source="local", name="wp-core-v1"),
+        dataset=DatasetConfig(source="local", name="wp-core"),
         model=ModelConfig(name="test-model"),
         grader=GraderConfig(kind="cli"),
         run=RunConfig(),
diff --git a/python/tests/test_execution_dataset.py b/python/tests/test_execution_dataset.py
index 3c54163..37996a6 100644
--- a/python/tests/test_execution_dataset.py
+++ b/python/tests/test_execution_dataset.py
@@ -7,7 +7,7 @@
 import orjson
 
 PROJECT_ROOT = Path(__file__).resolve().parents[2]
-EXECUTION_DIR = PROJECT_ROOT / "datasets" / "suites" / "wp-core-v1" / "execution"
+EXECUTION_DIR = PROJECT_ROOT / "datasets" / "suites" / "wp-core" / "execution"
 
 SUPPORTED_ASSERTIONS = {
     "class_exists",
diff --git a/python/tests/test_execution_isolation.py b/python/tests/test_execution_isolation.py
index dd28914..b18090d 100644
--- a/python/tests/test_execution_isolation.py
+++ b/python/tests/test_execution_isolation.py
@@ -23,7 +23,7 @@
 def _execution_test(test_id: str) -> ExecutionTest:
     return ExecutionTest(
         id=test_id,
-        suite="wp-core-v1",
+        suite="wp-core",
         prompt="Prompt",
         expected_behavior="expected",
         category="general",
@@ -47,7 +47,7 @@ def _passing_result() -> ExecutionResult:
 
 def _config(tmp_path: Path, **run_overrides: Any) -> HarnessConfig:
     return HarnessConfig(
-        dataset=DatasetConfig(source="local", name="wp-core-v1"),
+        dataset=DatasetConfig(source="local", name="wp-core"),
         model=ModelConfig(name="test-model"),
         grader=GraderConfig(kind="cli"),
         run=RunConfig(**run_overrides),
@@ -136,7 +136,7 @@ def test_multi_model_runner_resets_between_models(
         lambda dataset: tests,
     )
     config = HarnessConfig(
-        dataset=DatasetConfig(source="local", name="wp-core-v1"),
+        dataset=DatasetConfig(source="local", name="wp-core"),
         models=[ModelConfig(name="model-a"), ModelConfig(name="model-b")],
         grader=GraderConfig(kind="cli"),
         run=RunConfig(),
diff --git a/python/tests/test_execution_prompt.py b/python/tests/test_execution_prompt.py
index 8eacd64..412430c 100644
--- a/python/tests/test_execution_prompt.py
+++ b/python/tests/test_execution_prompt.py
@@ -12,7 +12,7 @@
 def _test(artifact_kind: str = "php_snippet") -> ExecutionTest:
     return ExecutionTest(
         id="e-x-001",
-        suite="wp-core-v1",
+        suite="wp-core",
         prompt="Register a thing.",
         expected_behavior="expected",
         category="hooks",
diff --git a/python/tests/test_exploit_audit.py b/python/tests/test_exploit_audit.py
index a29b185..98f2d14 100644
--- a/python/tests/test_exploit_audit.py
+++ b/python/tests/test_exploit_audit.py
@@ -23,7 +23,7 @@
 def _execution_test(test_id: str, test_function: str | None = "wpbp_x( int $n ): int") -> ExecutionTest:
     return ExecutionTest(
         id=test_id,
-        suite="wp-core-v1",
+        suite="wp-core",
         prompt="Prompt",
         expected_behavior="expected",
         category="general",
@@ -94,7 +94,7 @@ def execute_artifact(self, artifact: Any, verification_spec: dict) -> ExecutionR
 
 def _config(tmp_path: Path, **run_overrides: Any) -> HarnessConfig:
     return HarnessConfig(
-        dataset=DatasetConfig(source="local", name="wp-core-v1"),
+        dataset=DatasetConfig(source="local", name="wp-core"),
         model=ModelConfig(name="test-model"),
         grader=GraderConfig(kind="cli"),
         run=RunConfig(check_exploits=True, **run_overrides),
diff --git a/python/tests/test_reference_solution_mode.py b/python/tests/test_reference_solution_mode.py
index d9c14a6..d02a7ea 100644
--- a/python/tests/test_reference_solution_mode.py
+++ b/python/tests/test_reference_solution_mode.py
@@ -21,7 +21,7 @@
 
 def _config(tmp_path: Path, test_ids: list[str] | None = None) -> HarnessConfig:
     return HarnessConfig(
-        dataset=DatasetConfig(source="local", name="wp-core-v1"),
+        dataset=DatasetConfig(source="local", name="wp-core"),
         model=ModelConfig(name="test-model"),
         grader=GraderConfig(kind="cli"),
         run=RunConfig(
@@ -38,7 +38,7 @@ def _config(tmp_path: Path, test_ids: list[str] | None = None) -> HarnessConfig:
 def _execution_test(test_id: str = "e-one") -> ExecutionTest:
     return ExecutionTest(
         id=test_id,
-        suite="wp-core-v1",
+        suite="wp-core",
         prompt="Prompt",
         expected_behavior="Reviewer contract: expected",
         category="general",
diff --git a/python/tests/test_result_schema.py b/python/tests/test_result_schema.py
index a2eb890..1266e5c 100644
--- a/python/tests/test_result_schema.py
+++ b/python/tests/test_result_schema.py
@@ -25,7 +25,7 @@
 def _execution_test(test_id: str = "e-one") -> ExecutionTest:
     return ExecutionTest(
         id=test_id,
-        suite="wp-core-v1",
+        suite="wp-core",
         prompt="Prompt",
         expected_behavior="expected",
         category="hooks",
@@ -60,7 +60,7 @@ def _key_paths(record: dict[str, Any], prefix: str = "") -> set[str]:
 
 def _single_model_records(monkeypatch: pytest.MonkeyPatch, tmp_path: Path) -> list:
     config = HarnessConfig(
-        dataset=DatasetConfig(source="local", name="wp-core-v1"),
+        dataset=DatasetConfig(source="local", name="wp-core"),
         model=ModelConfig(name="test-model"),
         grader=GraderConfig(kind="cli"),
         run=RunConfig(),
@@ -81,7 +81,7 @@ def _single_model_records(monkeypatch: pytest.MonkeyPatch, tmp_path: Path) -> li
 
 def _multi_model_records(monkeypatch: pytest.MonkeyPatch, tmp_path: Path) -> list:
     config = HarnessConfig(
-        dataset=DatasetConfig(source="local", name="wp-core-v1"),
+        dataset=DatasetConfig(source="local", name="wp-core"),
         models=[ModelConfig(name="model-a")],
         grader=GraderConfig(kind="cli"),
         run=RunConfig(),
@@ -154,7 +154,7 @@ def test_reference_solution_record_uses_canonical_schema(
     tmp_path: Path,
 ) -> None:
     config = HarnessConfig(
-        dataset=DatasetConfig(source="local", name="wp-core-v1"),
+        dataset=DatasetConfig(source="local", name="wp-core"),
         model=ModelConfig(name="test-model"),
         grader=GraderConfig(kind="cli"),
         run=RunConfig(check_reference_solution=True),
@@ -191,7 +191,7 @@ def test_jsonl_records_match_json_results(
     tmp_path: Path,
 ) -> None:
     config = HarnessConfig(
-        dataset=DatasetConfig(source="local", name="wp-core-v1"),
+        dataset=DatasetConfig(source="local", name="wp-core"),
         model=ModelConfig(name="test-model"),
         grader=GraderConfig(kind="cli"),
         run=RunConfig(),
@@ -226,7 +226,7 @@ def test_records_are_sorted_for_stable_output(
     tmp_path: Path,
 ) -> None:
     config = HarnessConfig(
-        dataset=DatasetConfig(source="local", name="wp-core-v1"),
+        dataset=DatasetConfig(source="local", name="wp-core"),
         model=ModelConfig(name="test-model"),
         grader=GraderConfig(kind="cli"),
         run=RunConfig(),
diff --git a/python/tests/test_result_streaming.py b/python/tests/test_result_streaming.py
index 43fdf70..6d26e45 100644
--- a/python/tests/test_result_streaming.py
+++ b/python/tests/test_result_streaming.py
@@ -102,7 +102,7 @@ def _multi_model_config(tmp_path: Path, names: list[str]):
     )
 
     return HarnessConfig(
-        dataset=DatasetConfig(source="local", name="wp-core-v1"),
+        dataset=DatasetConfig(source="local", name="wp-core"),
         models=[ModelConfig(name=name) for name in names],
         grader=GraderConfig(kind="cli"),
         run=RunConfig(),
@@ -121,7 +121,7 @@ def _run_multi_model(monkeypatch, tmp_path: Path, names: list[str]):
 
     test = ExecutionTest(
         id="e-one",
-        suite="wp-core-v1",
+        suite="wp-core",
         prompt="Prompt",
         expected_behavior="expected",
         category="hooks",
diff --git a/python/tests/test_scoring.py b/python/tests/test_scoring.py
index fcd9c44..b3e3fe3 100644
--- a/python/tests/test_scoring.py
+++ b/python/tests/test_scoring.py
@@ -14,7 +14,7 @@ def _make_test(
 ) -> ExecutionTest:
     return ExecutionTest(
         id="e-test-001",
-        suite="wp-core-v1",
+        suite="wp-core",
         prompt="Do something.",
         expected_behavior="Reviewer contract: does something observable.",
         category="hooks",
diff --git a/python/tests/test_selection.py b/python/tests/test_selection.py
index bcfc01d..c5196cc 100644
--- a/python/tests/test_selection.py
+++ b/python/tests/test_selection.py
@@ -12,7 +12,7 @@
 def _test(test_id: str, category: str) -> ExecutionTest:
     return ExecutionTest(
         id=test_id,
-        suite="wp-core-v1",
+        suite="wp-core",
         prompt="Prompt",
         expected_behavior="expected",
         category=category,
@@ -176,7 +176,7 @@ def test_category_filtering_happens_before_limit_selection() -> None:
 def test_categories_flow_from_harness_config_to_run_selection() -> None:
     config = HarnessConfig.model_validate(
         {
-            "dataset": {"source": "local", "name": "wp-core-v1"},
+            "dataset": {"source": "local", "name": "wp-core"},
             "run": {"categories": ["bbb"]},
         }
     )
@@ -189,7 +189,7 @@ def test_categories_flow_from_harness_config_to_run_selection() -> None:
 def test_unknown_category_raises_value_error_naming_category() -> None:
     config = HarnessConfig.model_validate(
         {
-            "dataset": {"source": "local", "name": "wp-core-v1"},
+            "dataset": {"source": "local", "name": "wp-core"},
             "run": {"categories": ["not-a-category"]},
         }
     )
diff --git a/python/tests/test_shell_runtime_integration.py b/python/tests/test_shell_runtime_integration.py
index a87751a..c09e398 100644
--- a/python/tests/test_shell_runtime_integration.py
+++ b/python/tests/test_shell_runtime_integration.py
@@ -74,7 +74,7 @@ def test_role_changes_are_reloaded_before_assertions(environment) -> None:
 
 
 def test_capability_workflow_accepts_explicit_denial_as_well_as_removal(environment) -> None:
-    test = next(t for t in load_tests(DatasetConfig(source="local", name="wp-cli-v1")) if t.id == "e-cli-users-002")
+    test = next(t for t in load_tests(DatasetConfig(source="local", name="wp-cli")) if t.id == "e-cli-users-002")
     assert test.reference_solution is not None
     script = test.reference_solution.replace(
         "wp cap remove cli-contributor publish_posts", "wp cap add cli-contributor publish_posts --grant=false",
@@ -156,7 +156,7 @@ def test_shell_limits_remove_candidate_and_descendant_processes(environment, scr
 def test_slow_script_stops_inside_the_shell_and_keeps_its_output(environment, monkeypatch) -> None:
     grader = GraderConfig(timeout_seconds=10)
     monkeypatch.setattr(environment.config, "timeout_seconds", grader.timeout_seconds)
-    test = load_tests(DatasetConfig(source="local", name="wp-cli-v1"))[0]
+    test = load_tests(DatasetConfig(source="local", name="wp-cli"))[0]
     test.runtime_checks = {"assertions": [{"type": "custom_assertion", "code": "return true;"}]}
     environment.reset()
     result = environment.execute_artifact(
@@ -201,7 +201,7 @@ def test_rewriting_a_fixture_php_file_in_the_same_second_scores_zero(environment
 
 @pytest.mark.parametrize("test_id", ["e-cli-maintenance-001", "e-cli-maintenance-002"])
 def test_fixture_plugins_do_not_leak_into_the_next_test(environment, test_id) -> None:
-    test = next(t for t in load_tests(DatasetConfig(source="local", name="wp-cli-v1")) if t.id == test_id)
+    test = next(t for t in load_tests(DatasetConfig(source="local", name="wp-cli")) if t.id == test_id)
     artifact = Artifact(kind="wp_cli_shell", code=test.reference_solution or "")
     spec = _build_verification_spec(test, HarnessConfig())
     payload = base64.b64encode(json.dumps({**artifact.payload_fields(), **spec}).encode()).decode()
diff --git a/python/tests/test_skill_variants.py b/python/tests/test_skill_variants.py
index 9815df7..512686e 100644
--- a/python/tests/test_skill_variants.py
+++ b/python/tests/test_skill_variants.py
@@ -29,7 +29,7 @@
 def _execution_test(test_id: str = "e-one") -> ExecutionTest:
     return ExecutionTest(
         id=test_id,
-        suite="wp-core-v1",
+        suite="wp-core",
         prompt="Prompt",
         expected_behavior="expected",
         category="hooks",
@@ -74,7 +74,7 @@ def _skill_dir(tmp_path: Path) -> Path:
 
 def _config(tmp_path: Path, skill_path: Path) -> HarnessConfig:
     return HarnessConfig(
-        dataset=DatasetConfig(source="local", name="wp-core-v1"),
+        dataset=DatasetConfig(source="local", name="wp-core"),
         models=[ModelConfig(name="model-a")],
         grader=GraderConfig(kind="cli"),
         run=RunConfig(),
diff --git a/python/tests/test_usage_capture.py b/python/tests/test_usage_capture.py
index aacf299..90561e5 100644
--- a/python/tests/test_usage_capture.py
+++ b/python/tests/test_usage_capture.py
@@ -135,7 +135,7 @@ def test_result_record_contains_usage_and_model_call(
 ) -> None:
     test = ExecutionTest(
         id="e-one",
-        suite="wp-core-v1",
+        suite="wp-core",
         prompt="Prompt",
         expected_behavior="expected",
         category="hooks",
@@ -147,7 +147,7 @@ def test_result_record_contains_usage_and_model_call(
         metadata={},
     )
     config = HarnessConfig(
-        dataset=DatasetConfig(source="local", name="wp-core-v1"),
+        dataset=DatasetConfig(source="local", name="wp-core"),
         model=ModelConfig(name="gpt-4o-mini"),
         grader=GraderConfig(kind="cli"),
         run=RunConfig(),
diff --git a/python/tests/test_wp_cli_shell.py b/python/tests/test_wp_cli_shell.py
index c1d22fd..d18c3cb 100644
--- a/python/tests/test_wp_cli_shell.py
+++ b/python/tests/test_wp_cli_shell.py
@@ -45,21 +45,21 @@ def test_shell_artifact_rejects_invalid_size_or_content(script: str) -> None:
 
 @pytest.mark.parametrize("grader_timeout,shell_timeout", [(90, 85), (6, 1), (1, 1)])
 def test_shell_budget_ends_before_the_grader_timeout(grader_timeout, shell_timeout) -> None:
-    test = load_tests(DatasetConfig(source="local", name="wp-cli-v1"))[0]
+    test = load_tests(DatasetConfig(source="local", name="wp-cli"))[0]
     config = HarnessConfig(grader=GraderConfig(timeout_seconds=grader_timeout))
     assert _build_verification_spec(test, config)["timeout_seconds"] == shell_timeout
 
 
 @pytest.mark.parametrize("mode", ["model", "reference", "exploits"])
 def test_shell_scripts_reach_the_runtime_in_every_mode(monkeypatch, tmp_path, mode) -> None:
-    test = load_tests(DatasetConfig(source="local", name="wp-cli-v1"))[0]
+    test = load_tests(DatasetConfig(source="local", name="wp-cli"))[0]
     reference = test.reference_solution
     assert reference is not None
     monkeypatch.setattr("wp_bench.core.load_tests", lambda dataset: [test])
     config = HarnessConfig(
-        dataset=DatasetConfig(source="local", name="wp-cli-v1"),
+        dataset=DatasetConfig(source="local", name="wp-cli"),
         run=RunConfig(
-            suite="wp-cli-v1", check_reference_solution=mode == "reference",
+            suite="wp-cli", check_reference_solution=mode == "reference",
             check_exploits=mode == "exploits",
         ),
         output=OutputConfig(path=tmp_path / "result.json", jsonl_path=None),
@@ -87,7 +87,7 @@ def execute(artifact: Artifact, spec: dict) -> ExecutionResult:
 
 
 def test_shell_prompt_describes_the_actual_execution_contract() -> None:
-    test = load_tests(DatasetConfig(source="local", name="wp-cli-v1"))[0]
+    test = load_tests(DatasetConfig(source="local", name="wp-cli"))[0]
     prompt = BenchmarkRunner._render_execution_prompt(test)
     assert "Bash command or script" in prompt
     assert "SQLite" in prompt and "no network" in prompt
@@ -99,8 +99,8 @@ def test_shell_prompt_describes_the_actual_execution_contract() -> None:
 
 def test_cli_suite_flag_selects_the_local_cli_suite(tmp_path) -> None:
     path = tmp_path / "config.yaml"
-    path.write_text("dataset: {source: local, name: wp-core-v1}\n")
-    result = CliRunner().invoke(app, ["run", "--config", str(path), "--suite", "wp-cli-v1", "--dry-run"])
+    path.write_text("dataset: {source: local, name: wp-core}\n")
+    result = CliRunner().invoke(app, ["run", "--config", str(path), "--suite", "wp-cli", "--dry-run"])
     assert result.exit_code == 0, result.output
     assert "Execution tests: 27" in result.output
 
@@ -113,30 +113,30 @@ def load(name, **kwargs):
         return [
             {"id": suite, "suite": suite, "test_kind": "execution", "prompt": "Task",
              "artifact_kind": kind}
-            for suite, kind in [("wp-core-v1", "php_snippet"), ("wp-cli-v1", "wp_cli_shell")]
+            for suite, kind in [("wp-core", "php_snippet"), ("wp-cli", "wp_cli_shell")]
         ]
 
     monkeypatch.setattr("wp_bench.datasets.hf_load_dataset", load)
-    selected = load_tests(DatasetConfig(source="huggingface", name="WordPress/wp-bench-v1", suite="wp-cli-v1"))
+    selected = load_tests(DatasetConfig(source="huggingface", name="WordPress/wp-bench-v1", suite="wp-cli"))
     assert [t.artifact_kind for t in selected] == ["wp_cli_shell"]
     path = tmp_path / "config.yaml"
     path.write_text("dataset: {source: huggingface, name: WordPress/wp-bench-v1}\n")
-    result = CliRunner().invoke(app, ["run", "--config", str(path), "--suite", "wp-cli-v1", "--dry-run"])
+    result = CliRunner().invoke(app, ["run", "--config", str(path), "--suite", "wp-cli", "--dry-run"])
     assert result.exit_code == 0, result.output
     assert "Execution tests: 1" in result.output
     assert seen == ["WordPress/wp-bench-v1"] * 2
 
 
 @pytest.mark.parametrize("dataset,run,suite", [
-    ({}, {}, "wp-core-v1"),
-    ({}, {"suite": "wp-cli-v1"}, "wp-cli-v1"),
-    ({"suite": "wp-cli-v1"}, {}, "wp-cli-v1"),
-    ({"suite": "wp-cli-v1"}, {"suite": "wp-core-v1"}, "wp-cli-v1"),
-    ({"suite": "wp-core-v1"}, {"suite": "wp-cli-v1"}, "wp-core-v1"),
+    ({}, {}, "wp-core"),
+    ({}, {"suite": "wp-cli"}, "wp-cli"),
+    ({"suite": "wp-cli"}, {}, "wp-cli"),
+    ({"suite": "wp-cli"}, {"suite": "wp-core"}, "wp-cli"),
+    ({"suite": "wp-core"}, {"suite": "wp-cli"}, "wp-core"),
 ])
 def test_huggingface_loads_only_the_run_suite_unless_overridden(monkeypatch, dataset, run, suite) -> None:
     rows = [{"id": suite, "suite": suite, "test_kind": "execution", "prompt": "Task"}
-            for suite in ["wp-core-v1", "wp-cli-v1"]]
+            for suite in ["wp-core", "wp-cli"]]
     monkeypatch.setattr("wp_bench.datasets.hf_load_dataset", lambda name, **kwargs: rows)
     config = HarnessConfig.model_validate({"dataset": {"source": "huggingface", **dataset}, "run": run})
     assert [t.suite for t in load_tests(config.dataset)] == [suite]
@@ -144,17 +144,17 @@ def test_huggingface_loads_only_the_run_suite_unless_overridden(monkeypatch, dat
 
 
 def test_cli_suite_has_workflows_with_behavioral_and_exploit_checks() -> None:
-    tests = load_tests(DatasetConfig(source="local", name="wp-cli-v1"))
+    tests = load_tests(DatasetConfig(source="local", name="wp-cli"))
     assert Counter(t.category for t in tests) == {
         "cli-content": 3, "cli-taxonomy-meta": 4, "cli-users": 5,
         "cli-options": 4, "cli-reporting": 3, "cli-maintenance": 6,
         "cli-migration": 2,
     }
-    core = load_tests(DatasetConfig(source="local", name="wp-core-v1"))
+    core = load_tests(DatasetConfig(source="local", name="wp-core"))
     assert len(core) == 350
     assert len({t.id for t in core + tests}) == 377
     for test in tests:
-        assert test.suite == "wp-cli-v1" and test.artifact_kind == "wp_cli_shell"
+        assert test.suite == "wp-cli" and test.artifact_kind == "wp_cli_shell"
         assert test.test_function is None
         assert test.reference_solution and "wp " in test.reference_solution
         assert test.exploit_solutions and all(s.strip() for s in test.exploit_solutions)
@@ -175,10 +175,10 @@ def test_cli_export_preserves_shell_contract_and_excludes_exploits() -> None:
     assert spec and spec.loader
     module = importlib.util.module_from_spec(spec)
     spec.loader.exec_module(module)
-    rows = module.load_suite("wp-cli-v1")
+    rows = module.load_suite("wp-cli")
     assert len(rows) == 27
     assert all(r["artifact_kind"] == "wp_cli_shell" and "exploit_solutions" not in r for r in rows)
-    local = {t.id: t for t in load_tests(DatasetConfig(source="local", name="wp-cli-v1"))}
+    local = {t.id: t for t in load_tests(DatasetConfig(source="local", name="wp-cli"))}
     for row in rows:
         assert row["reference_solution"] == local[row["id"]].reference_solution
         assert json.loads(row["runtime_checks"]) == local[row["id"]].runtime_checks
diff --git a/python/wp_bench/benchmark.py b/python/wp_bench/benchmark.py
new file mode 100644
index 0000000..4aee71b
--- /dev/null
+++ b/python/wp_bench/benchmark.py
@@ -0,0 +1,298 @@
+"""Frozen benchmark releases shared by the harness and leaderboard consumers.
+
+A public version identifies the entire evaluation. Suite names, result schema
+versions, and result update dates are separate provenance, never substitutes.
+"""
+from __future__ import annotations
+
+import hashlib
+import json
+import re
+from dataclasses import dataclass
+from pathlib import Path
+from typing import Any
+
+from .config import GraderConfig, HarnessConfig, ModelConfig
+from .datasets import ExecutionTest
+from .scoring import SCORING_VERSION
+
+PROJECT_ROOT = Path(__file__).resolve().parents[2]
+CONTRACT_SCHEMA_VERSION = "1.0"
+VERSION_PATTERN = re.compile(r"v[1-9][0-9]*")
+# Conservative: changes to these evaluation implementations need a new release.
+# CLI/output/record serialization changes alone do not change the evaluation.
+HARNESS_INPUTS = (
+    "artifacts.py", "config.py", "utils.py", "core.py", "datasets.py", "environment.py", "models.py",
+    "scoring.py", "selection.py", "skills.py", "exploits.py",
+)
+
+
+def digest(value: Any) -> str:
+    """Hash canonical JSON, independent of object key order."""
+    body = json.dumps(value, sort_keys=True, separators=(",", ":"), allow_nan=False)
+    return hashlib.sha256(body.encode()).hexdigest()
+
+
+def read_json(path: Path) -> dict[str, Any]:
+    """Read an object with a useful error for a broken/missing manifest."""
+    try:
+        value = json.loads(path.read_text())
+    except (OSError, ValueError) as exc:
+        raise ValueError(f"Cannot read benchmark manifest {path}: {exc}") from exc
+    if not isinstance(value, dict):
+        raise TypeError(f"Benchmark manifest must be an object: {path}")
+    return value
+
+
+def task_digest(test: ExecutionTest) -> str:
+    """Evaluation inputs that survive the local/Hugging Face round trip.
+
+    ``test.suite`` is a source-file ID locally and a dataset ID on the Hub;
+    the enclosing run's dataset ID supplies the component identity instead.
+    Reference solutions, author QA and editorial metadata are not model input.
+    Raw dataset file hashes separately freeze *all* dataset changes.
+    """
+    return digest({name: getattr(test, name) for name in (
+        "id", "prompt", "requirements", "category", "test_function",
+        "static_checks", "runtime_checks", "artifact_kind",
+    )})
+
+
+def evaluation_snapshot(root: Path = PROJECT_ROOT) -> dict[str, Any]:
+    """Capture every component and its complete test membership."""
+    from .datasets import _parse_execution_suite
+
+    definition = read_json(root / "benchmark/definition.json")
+    if definition.get("schema_version") != CONTRACT_SCHEMA_VERSION:
+        raise ValueError("Unsupported benchmark definition schema")
+    if definition.get("scoring_version") != SCORING_VERSION:
+        raise ValueError("Working definition must use the current scoring formula")
+    protocol = definition.get("protocol", {})
+    ModelConfig(**protocol.get("model", {}))
+    GraderConfig(**protocol.get("grader", {}))
+    repeats = protocol.get("repeats")
+    if not isinstance(repeats, int) or isinstance(repeats, bool) or repeats < 1:
+        raise ValueError("Benchmark protocol must require a positive number of repeats")
+    approved = definition.get("runtime_image_ids", [])
+    if not approved or len(approved) != len(set(approved)):
+        raise ValueError("Benchmark must declare distinct approved runtime images")
+    labels = [category["id"] for category in definition.get("categories", [])]
+    if not labels or len(labels) != len(set(labels)):
+        raise ValueError("Broad category IDs must be nonempty and distinct")
+    suites = definition.get("suites", [])
+    if not suites or len(suites) != len(set(suites)):
+        raise ValueError("Benchmark must list distinct dataset suites")
+    if set(definition.get("category_mapping", {})) != set(suites):
+        raise ValueError("Category mappings must cover exactly the declared suites")
+    files: dict[str, str] = {}
+    components: dict[str, Any] = {}
+    actual_suites = sorted(p.name for p in (root / "datasets/suites").iterdir() if p.is_dir())
+    if sorted(suites) != actual_suites:
+        raise ValueError("Benchmark definition must include every dataset suite")
+    for suite in suites:
+        if not re.fullmatch(r"[A-Za-z0-9_-]+", suite):
+            raise ValueError(f"Invalid dataset suite: {suite}")
+        tests = []
+        for path in sorted((root / "datasets/suites" / suite / "execution").glob("*.json")):
+            files[path.relative_to(root).as_posix()] = hashlib.sha256(path.read_bytes()).hexdigest()
+            tests.extend(_parse_execution_suite(path))
+        ids = [test.id for test in tests]
+        if not ids or len(ids) != len(set(ids)):
+            raise ValueError(f"Suite {suite} must contain distinct tests")
+        components[suite] = {"task_count": len(tests), "tasks": {
+            test.id: {"sha256": task_digest(test), "category": test.category}
+            for test in sorted(tests, key=lambda test: test.id)
+        }}
+        # Keep public category membership exhaustive and unambiguous.
+        mapping = definition["category_mapping"].get(suite, {})
+        categories = {test.category for test in tests}
+        allowed_labels = set(labels)
+        if set(mapping) != categories or any(value not in allowed_labels for value in mapping.values()):
+            raise ValueError(f"Broad category mapping is incomplete for {suite}")
+    for path in sorted((root / "runtime").rglob("*")):
+        if path.is_file() and (path.suffix in {".php", ".sh"} or path.name == "Dockerfile"):
+            files[path.relative_to(root).as_posix()] = hashlib.sha256(path.read_bytes()).hexdigest()
+    for name in HARNESS_INPUTS:
+        path = root / "python/wp_bench" / name
+        files[path.relative_to(root).as_posix()] = hashlib.sha256(path.read_bytes()).hexdigest()
+    return {"definition": definition, "files": files, "components": components}
+
+
+def load_release(version: str | None = None, root: Path = PROJECT_ROOT) -> dict[str, Any]:
+    """Read a frozen public release; validate its content-addressed identity."""
+    catalog = read_json(root / "benchmark/releases.json")
+    version = version or catalog["current"]
+    if not isinstance(version, str) or not VERSION_PATTERN.fullmatch(version):
+        raise ValueError("Benchmark versions must be v1, v2, ...")
+    entry = next((entry for entry in catalog["releases"] if entry["version"] == version), None)
+    if entry is None:
+        raise ValueError(f"Unknown benchmark version: {version}")
+    manifest = read_json(root / "benchmark/releases" / f"{version}.json")
+    if (manifest.get("schema_version") != CONTRACT_SCHEMA_VERSION
+            or manifest.get("version") != version
+            or manifest.get("fingerprint") != digest(manifest.get("evaluation"))
+            or entry.get("fingerprint") != manifest["fingerprint"]):
+        raise ValueError(f"Invalid benchmark release manifest: {version}")
+    return manifest
+
+
+def freeze_release(version: str, root: Path = PROJECT_ROOT) -> dict[str, Any]:
+    """Append a release, refusing to replace a previously published identity."""
+    if not VERSION_PATTERN.fullmatch(version):
+        raise ValueError("Benchmark versions must be v1, v2, ...")
+    path = root / "benchmark/releases" / f"{version}.json"
+    if path.exists():
+        raise ValueError(f"Benchmark {version} already exists; published releases are immutable")
+    catalog_path = root / "benchmark/releases.json"
+    catalog = read_json(catalog_path) if catalog_path.exists() else {
+        "schema_version": CONTRACT_SCHEMA_VERSION, "releases": [],
+    }
+    expected = max((int(entry["version"][1:]) for entry in catalog["releases"]), default=0) + 1
+    if version != f"v{expected}":
+        raise ValueError(f"Next benchmark release must be v{expected}")
+    evaluation = evaluation_snapshot(root)
+    if any(entry["fingerprint"] == digest(evaluation) for entry in catalog["releases"]):
+        raise ValueError("Evaluation is unchanged; export updated results under the existing version")
+    manifest = {"schema_version": CONTRACT_SCHEMA_VERSION, "version": version,
+                "fingerprint": digest(evaluation), "evaluation": evaluation}
+    path.parent.mkdir(parents=True, exist_ok=True)
+    # Exclusive create protects old releases even if two release jobs race.
+    with path.open("x") as handle:
+        handle.write(json.dumps(manifest, indent=2) + "\n")
+    catalog["releases"].append({"version": version, "path": f"releases/{version}.json",
+                                "fingerprint": manifest["fingerprint"]})
+    catalog["current"] = version
+    catalog_path.write_text(json.dumps(catalog, indent=2) + "\n")
+    return manifest
+
+
+@dataclass
+class BenchmarkIdentity:
+    """Capture once before model calls so later disk edits cannot relabel a run."""
+
+    manifest: dict[str, Any] | None
+    fingerprint: str | None
+    issues: list[str]
+    suite: str
+    tasks: dict[str, str]
+
+    def record_info(self) -> dict[str, Any]:
+        """Small identity repeated on streamed records, including partial runs."""
+        return {"version": self.manifest["version"] if self.manifest and not self.issues else None,
+                "fingerprint": self.fingerprint, "suite": self.suite,
+                "status": "released" if self.manifest and not self.issues else "unreleased"}
+
+    def bind_runtime(self, image_id: str | None, *, required: bool = False) -> None:
+        """Verify the actual image before generation, rather than trusting a tag."""
+        if self.manifest and image_id not in self.manifest["evaluation"]["definition"]["runtime_image_ids"]:
+            self.issues.append("Runtime image is not approved for this release")
+        if required and self.issues:
+            raise ValueError("Cannot run as a public benchmark release: " + "; ".join(self.issues))
+
+    def metadata(self, records: list[dict[str, Any]]) -> dict[str, Any]:
+        """Report exact coverage without promoting partial or errored runs."""
+        info = self.record_info()
+        expected = self.manifest["evaluation"]["components"].get(self.suite, {}).get("tasks", {}) if self.manifest else {}
+        ids = [record["test_id"] for record in records]
+        complete = bool(expected) and len(ids) == len(set(ids)) and set(ids) == set(expected)
+        return {"schema_version": CONTRACT_SCHEMA_VERSION, **info,
+                "based_on": self.manifest["version"] if self.manifest else None,
+                "issues": self.issues, "full_suite": complete,
+                "expected_task_count": len(expected), "observed_task_count": len(ids)}
+
+
+def capture_identity(config: HarnessConfig, tests: list[ExecutionTest], *,
+                     root: Path = PROJECT_ROOT) -> BenchmarkIdentity:
+    """Identify source/test changes; unknown inputs never inherit a public label."""
+    issues: list[str] = []
+    tasks = {test.id: task_digest(test) for test in tests}
+    manifest = None
+    fingerprint = None
+    try:
+        manifest = load_release(config.benchmark.version, root)
+        snapshot = evaluation_snapshot(root)
+        fingerprint = digest(snapshot)
+        if fingerprint != manifest["fingerprint"]:
+            issues.append("Evaluation sources differ from the frozen release")
+        component = manifest["evaluation"]["components"].get(config.run.suite, {})
+        expected = component.get("tasks", {})
+        if not expected or len(tasks) != len(tests) or any(
+            test_id not in expected or value != expected[test_id]["sha256"]
+            for test_id, value in tasks.items()
+        ):
+            issues.append("Loaded tasks differ from the frozen release")
+        protocol = manifest["evaluation"]["definition"]["protocol"]
+        if config.run.skip_static or config.run.skip_runtime or config.run.continue_on_error:
+            issues.append("Diagnostic run settings are enabled")
+        if config.run.execution_isolation != "reset_per_test":
+            issues.append("Per-test isolation is disabled")
+        if any(getattr(config.grader, key) != value for key, value in protocol["grader"].items()):
+            issues.append("Grader settings differ from the release protocol")
+        if not config.run.check_reference_solution and not config.run.check_exploits:
+            for model in config.get_models():
+                if any(getattr(model, key) != value for key, value in protocol["model"].items()):
+                    issues.append(f"Model settings differ from the release protocol: {model.name}")
+        if SCORING_VERSION != manifest["evaluation"]["definition"]["scoring_version"]:
+            issues.append("Scoring version differs from the frozen release")
+    except (ValueError, TypeError) as exc:
+        issues.append(str(exc))
+    if config.benchmark.version and issues:
+        raise ValueError("Cannot run as " + config.benchmark.version + ": " + "; ".join(issues))
+    return BenchmarkIdentity(manifest, fingerprint, issues, config.run.suite, tasks)
+
+
+def publication_issues(payload: dict[str, Any], manifest: dict[str, Any]) -> list[str]:
+    """Validate a full-suite model result before any leaderboard ingestion.
+
+    Receives a single-model payload. Multi-model callers adapt each models[]
+    entry into this shape. No score is recomputed from a partial result.
+    """
+    metadata = payload.get("metadata", {})
+    identity = metadata.get("benchmark") or {}
+    records = payload.get("results", [])
+    evaluation = manifest["evaluation"]
+    definition = evaluation["definition"]
+    issues = []
+    if (identity.get("version") != manifest["version"]
+            or identity.get("fingerprint") != manifest["fingerprint"]
+            or identity.get("status") != "released"):
+        issues.append("Missing or incompatible benchmark identity")
+    suite = metadata.get("suite")
+    expected = evaluation["components"].get(suite, {}).get("tasks", {})
+    ids = [record.get("test_id") for record in records]
+    if not expected or len(ids) != len(set(ids)) or set(ids) != set(expected):
+        issues.append("Incomplete, duplicate, or unknown suite tasks")
+    if metadata.get("scoring_version") != definition["scoring_version"]:
+        issues.append("Scoring version mismatch")
+    if metadata.get("runtime_image_id") not in definition["runtime_image_ids"]:
+        issues.append("Runtime image is not approved for this release")
+    grader = metadata.get("grader", {})
+    if any(grader.get(key) != value for key, value in definition["protocol"]["grader"].items()):
+        issues.append("Grader settings differ from the release protocol")
+    if metadata.get("runtime_isolation") != "reset_per_test":
+        issues.append("Run did not use per-test isolation")
+    run_id = metadata.get("run_id")
+    if not isinstance(run_id, str) or not run_id:
+        issues.append("Missing run identity")
+    for record in records:
+        if record.get("run_id") != run_id:
+            issues.append("Record run identity mismatch")
+        task = expected.get(record.get("test_id"), {})
+        if (record.get("category") != task.get("category")
+                or (record.get("benchmark") or {}).get("task_sha256") != task.get("sha256")):
+            issues.append("Record task fingerprint mismatch")
+        model = record.get("model") or {}
+        if (record.get("mode") != "model" or record.get("error") is not None
+                or (record.get("model_call") or {}).get("temperature_fallback", False)
+                or record.get("variant", {}).get("kind") != "none"
+                or not isinstance(record.get("scores", {}).get("execution_pass"), bool)):
+            issues.append("Diagnostic, errored, or ungraded record")
+        if any(model.get(key) != value for key, value in definition["protocol"]["model"].items()):
+            issues.append("Model settings differ from the release protocol")
+        record_identity = record.get("benchmark") or {}
+        if (record_identity.get("version") != manifest["version"]
+                or record_identity.get("fingerprint") != manifest["fingerprint"]
+                or record_identity.get("suite") != suite
+                or record_identity.get("status") != "released"):
+            issues.append("Record benchmark identity mismatch")
+    return list(dict.fromkeys(issues))
diff --git a/python/wp_bench/benchmark_cli.py b/python/wp_bench/benchmark_cli.py
new file mode 100644
index 0000000..805921d
--- /dev/null
+++ b/python/wp_bench/benchmark_cli.py
@@ -0,0 +1,185 @@
+"""Release and leaderboard commands; no provider calls or runtime startup."""
+from __future__ import annotations
+
+import json
+import subprocess
+from datetime import datetime, timezone
+from pathlib import Path
+from typing import Any
+
+import typer
+
+from .benchmark import (
+    CONTRACT_SCHEMA_VERSION,
+    PROJECT_ROOT,
+    digest,
+    evaluation_snapshot,
+    freeze_release,
+    load_release,
+    publication_issues,
+    read_json,
+)
+
+app = typer.Typer(help="Inspect and freeze public benchmark releases; export validated results.")
+
+
+def check_releases(root: Path = PROJECT_ROOT, base_ref: str | None = None) -> dict[str, Any]:
+    """Validate the catalog and protect all releases already present upstream."""
+    catalog = read_json(root / "benchmark/releases.json")
+    if catalog.get("schema_version") != CONTRACT_SCHEMA_VERSION:
+        raise ValueError("Unsupported benchmark catalog schema")
+    entries = catalog.get("releases", [])
+    versions = [entry["version"] for entry in entries]
+    if not versions or versions != [f"v{number}" for number in range(1, len(versions) + 1)]:
+        raise ValueError("Release catalog must contain v1, v2, ... exactly once, in order")
+    if catalog.get("current") != versions[-1]:
+        raise ValueError("Current benchmark must be the latest frozen release")
+    if {path.stem for path in (root / "benchmark/releases").glob("*.json")} != set(versions):
+        raise ValueError("Release files and catalog differ")
+    for entry in entries:
+        if entry.get("path") != f"releases/{entry['version']}.json":
+            raise ValueError("Unexpected release manifest path")
+        load_release(entry["version"], root)
+    latest = load_release(root=root)
+    matches = digest(evaluation_snapshot(root)) == latest["fingerprint"]
+    if base_ref:
+        def git(*args: str) -> bytes:
+            result = subprocess.run(["git", *args], cwd=root, capture_output=True, timeout=30, check=False)
+            if result.returncode:
+                raise ValueError("Cannot inspect benchmark release history: " + result.stderr.decode().strip())
+            return result.stdout
+        base_files = git("ls-tree", "-r", "--name-only", base_ref, "--", "benchmark/releases").decode().splitlines()
+        for relative in base_files:
+            path = root / relative
+            if not path.is_file() or path.read_bytes() != git("show", f"{base_ref}:{relative}"):
+                raise ValueError(f"Published benchmark manifest changed or disappeared: {relative}")
+        if f"benchmark/releases/{latest['version']}.json" not in base_files and not matches:
+            raise ValueError("New release does not describe the current evaluation; freeze again before publishing")
+    return {"current": latest["version"], "fingerprint": latest["fingerprint"],
+            "working_tree": "released" if matches else "unreleased", "releases": versions}
+
+
+def export_results(paths: list[Path], output: Path, *, version: str | None = None,
+                   root: Path = PROJECT_ROOT) -> Path:
+    """Publish an immutable, version-scoped scored-answer bundle for websites.
+
+    Require complete suites and the same number of independent runs for every
+    model. Preserve old snapshots. Consumers derive displays from these rows;
+    no provider's self-reported aggregate can replace the actual task results.
+    """
+    manifest = load_release(version, root)
+    definition = manifest["evaluation"]["definition"]
+    models: dict[str, list[dict[str, Any]]] = {}
+    runs: dict[tuple[str, str], set[str]] = {}
+    used_runs: set[str] = set()
+    inputs: list[str] = []
+    for path in paths:
+        if path.name.endswith((".partial", ".jsonl")):
+            raise ValueError(f"Expected completed results JSON, not a stream: {path}")
+        payload = read_json(path)
+        inputs.append(digest(payload))
+        entries = [payload]
+        if "models" in payload:
+            entries = [{"metadata": {**payload["metadata"], "run_id": result["run_id"],
+                                      "benchmark": result["benchmark"]}, "results": result["results"]}
+                       for result in payload["models"].values()]
+        for result in entries:
+            issues = publication_issues(result, manifest)
+            if issues:
+                raise ValueError(f"Cannot publish {path.name}: {'; '.join(issues)}")
+            records = result["results"]
+            names = {record["model"]["name"] for record in records}
+            if len(names) != 1:
+                raise ValueError("A model run must contain exactly one model identity")
+            name = names.pop()
+            metadata = result["metadata"]
+            suite, run_id = metadata["suite"], metadata["run_id"]
+            if run_id in used_runs:
+                raise ValueError("Duplicate run identity; the same run cannot count as another repeat")
+            used_runs.add(run_id)
+            runs.setdefault((name, suite), set()).add(run_id)
+            answers = models.setdefault(name, [])
+            for record in records:
+                task = manifest["evaluation"]["components"][suite]["tasks"][record["test_id"]]
+                answers.append({"suite": suite, "test_id": record["test_id"], "run_id": run_id,
+                                "category": definition["category_mapping"][suite][task["category"]],
+                                "passed": record["scores"]["execution_pass"],
+                                "usage": {key: (record.get("usage") or {}).get(key) for key in (
+                                    "prompt_tokens", "completion_tokens", "total_tokens", "cost_usd", "latency_ms")}})
+    if not models:
+        raise ValueError("No completed model results supplied")
+    repeats = definition["protocol"]["repeats"]
+    for name in models:
+        for suite in definition["suites"]:
+            if len(runs.get((name, suite), set())) != repeats:
+                raise ValueError(f"{name} needs exactly {repeats} independent complete runs of {suite}")
+    content: dict[str, Any] = {"schema_version": CONTRACT_SCHEMA_VERSION,
+               "benchmark": {"version": manifest["version"], "fingerprint": manifest["fingerprint"]},
+               "categories": definition["categories"],
+               "models": [{"id": name, "answers": sorted(answers, key=lambda answer: (
+                   answer["suite"], answer["test_id"], answer["run_id"]))}
+                   for name, answers in sorted(models.items())], "input_sha256": sorted(inputs)}
+    fingerprint = digest(content)
+    snapshot_id = fingerprint[:16]
+    directory = output / manifest["version"]
+    destination = directory / f"{snapshot_id}.json"
+    catalog_path = output / "results.json"
+    catalog: dict[str, Any] = read_json(catalog_path) if catalog_path.exists() else {
+        "schema_version": CONTRACT_SCHEMA_VERSION, "benchmarks": {},
+    }
+    # Same exact inputs are idempotent, including the original update date.
+    if destination.exists():
+        existing = read_json(destination)
+        existing_content = {key: value for key, value in existing.items()
+                            if key not in {"snapshot_id", "fingerprint", "updated_at"}}
+        if existing.get("fingerprint") != fingerprint or digest(existing_content) != fingerprint:
+            raise ValueError("Existing result snapshot has a different fingerprint")
+    else:
+        snapshot = {**content, "snapshot_id": snapshot_id, "fingerprint": fingerprint,
+                    "updated_at": datetime.now(timezone.utc).isoformat()}
+        directory.mkdir(parents=True, exist_ok=True)
+        with destination.open("x") as handle:
+            handle.write(json.dumps(snapshot, indent=2, allow_nan=False) + "\n")
+    entry = catalog["benchmarks"].setdefault(manifest["version"], {"current": snapshot_id, "snapshots": []})
+    if snapshot_id not in [item["id"] for item in entry["snapshots"]]:
+        entry["snapshots"].append({"id": snapshot_id, "path": f"{manifest['version']}/{snapshot_id}.json"})
+    entry["current"] = snapshot_id
+    temporary = catalog_path.with_suffix(".json.tmp")
+    temporary.write_text(json.dumps(catalog, indent=2) + "\n")
+    temporary.replace(catalog_path)
+    return destination
+
+
+def _show(action: Any) -> None:
+    try:
+        result = action()
+    except (ValueError, KeyError, TypeError) as exc:
+        typer.echo(f"Error: {exc}", err=True)
+        raise typer.Exit(1) from exc
+    typer.echo(json.dumps(result, indent=2))
+
+
+@app.command("show")
+def show(version: str | None = typer.Option(None, help="Public release, such as v1")) -> None:
+    """Print the complete machine-readable release manifest."""
+    _show(lambda: load_release(version))
+
+
+@app.command("check")
+def check(base_ref: str | None = typer.Option(None, help="Git commit whose published releases must remain immutable")) -> None:
+    """Validate releases; report changed evaluation sources as unreleased."""
+    _show(lambda: check_releases(base_ref=base_ref))
+
+
+@app.command("freeze")
+def freeze(version: str = typer.Argument(..., help="Next public version, e.g. v2")) -> None:
+    """Freeze the entire current evaluation as the next immutable release."""
+    _show(lambda: {key: value for key, value in freeze_release(version).items() if key != "evaluation"})
+
+
+@app.command("export")
+def export(inputs: list[Path] = typer.Option(..., "--input", help="Completed result JSON; repeat for every suite and run"),
+           output: Path = typer.Option(..., "--output-dir", help="Archive root; keeps snapshots for every benchmark version"),
+           version: str | None = typer.Option(None, help="Public release, defaults to latest")) -> None:
+    """Validate results and write an immutable leaderboard data snapshot."""
+    _show(lambda: {"path": str(export_results(inputs, output, version=version))})
diff --git a/python/wp_bench/cli.py b/python/wp_bench/cli.py
index 45af891..5a226ac 100644
--- a/python/wp_bench/cli.py
+++ b/python/wp_bench/cli.py
@@ -9,6 +9,7 @@
 from dotenv import load_dotenv
 from rich.console import Console
 
+from .benchmark_cli import app as benchmark_app
 from .config import HarnessConfig, ModelConfig
 from .core import BenchmarkRunner, MultiModelRunner, select_run_tests
 from .datasets import ExecutionTest, filter_tests_by_ids, load_tests
@@ -20,6 +21,7 @@
 T = TypeVar("T")
 
 app = typer.Typer(add_completion=False)
+app.add_typer(benchmark_app, name="benchmark")
 console = Console()
 
 
diff --git a/python/wp_bench/config.py b/python/wp_bench/config.py
index a0ec196..df8ffc5 100644
--- a/python/wp_bench/config.py
+++ b/python/wp_bench/config.py
@@ -36,8 +36,8 @@ class StrictModel(BaseModel):
 
 
 class DatasetConfig(StrictModel):
-    source: Literal["huggingface", "local"] = "huggingface"
-    name: str = "WordPress/wp-bench-v1"
+    source: Literal["huggingface", "local"] = "local"
+    name: str = "wp-core"
     revision: str | None = None
     split: str = "test"
     cache_dir: Path | None = None
@@ -133,7 +133,7 @@ class RunConfig(StrictModel):
     #: exclusive-mode rules as a config file.
     model_config = ConfigDict(validate_assignment=True)
 
-    suite: str = "wp-core-v1"
+    suite: str = "wp-core"
     limit: int | None = Field(default=None, gt=0)
     test_ids: list[str] = Field(default_factory=list)
     categories: list[str] = Field(default_factory=list)
@@ -219,7 +219,23 @@ def _validate_only_requires_paths(self) -> SkillsConfig:
         return self
 
 
+class BenchmarkConfig(StrictModel):
+    #: Optional release assertion. When set, reject mismatched evaluation inputs
+    #: before spending on model calls. Unset runs still record their provenance.
+    version: str | None = None
+
+    @field_validator("version")
+    @classmethod
+    def _validate_version(cls, value: str | None) -> str | None:
+        import re
+
+        if value is not None and not re.fullmatch(r"v[1-9][0-9]*", value):
+            raise ValueError("benchmark.version must be v1, v2, ...")
+        return value
+
+
 class HarnessConfig(StrictModel):
+    benchmark: BenchmarkConfig = BenchmarkConfig()
     dataset: DatasetConfig = DatasetConfig()
     model: ModelConfig | None = None  # Single model (legacy)
     models: list[ModelConfig] | None = None  # Multiple models
diff --git a/python/wp_bench/core.py b/python/wp_bench/core.py
index 114052e..6d99041 100644
--- a/python/wp_bench/core.py
+++ b/python/wp_bench/core.py
@@ -7,6 +7,7 @@
 from concurrent.futures import CancelledError, ThreadPoolExecutor, as_completed
 from contextlib import contextmanager, suppress
 from typing import Any
+from uuid import uuid4
 
 from .artifacts import (
     Artifact,
@@ -14,6 +15,7 @@
     parse_artifact,
     render_artifact_instructions,
 )
+from .benchmark import BenchmarkIdentity, capture_identity
 from .config import HarnessConfig, ModelConfig
 from .datasets import (
     ExecutionTest,
@@ -88,6 +90,8 @@ class _ResultBookkeeping:
     """
 
     def _init_bookkeeping(self, stream: RecordStream) -> None:
+        self._run_id = uuid4().hex
+        self._benchmark_identity: BenchmarkIdentity | None = None
         self.aggregator = ScoreAggregator()
         self.usage_aggregator = UsageAggregator()
         self.records: list[dict[str, Any]] = []
@@ -97,6 +101,12 @@ def _init_bookkeeping(self, stream: RecordStream) -> None:
     def _on_result(self, result: dict[str, Any]) -> None:
         """Aggregate a finished record, keep it, and stream it to disk."""
         with self._lock:
+            result["run_id"] = self._run_id
+            if self._benchmark_identity is not None:
+                result["benchmark"] = {
+                    **self._benchmark_identity.record_info(),
+                    "task_sha256": self._benchmark_identity.tasks.get(result["test_id"]),
+                }
             if result.get("error") is None:
                 self.aggregator.add_execution(result["scores"])
             self.usage_aggregator.add(result.get("usage"))
@@ -500,11 +510,13 @@ def run(self) -> dict[str, Any]:
                 f"Dataset '{self.config.dataset.name}' contains no execution "
                 "tests. Check the dataset source and suite name."
             )
+        self._benchmark_identity = capture_identity(self.config, tests)
         if self.config.run.check_exploits:
             return self._run_exploit_audit(tests)
         reference_mode = self.config.run.check_reference_solution
         with _graded_run(self._stream, self.environment):
             self.environment.setup(capture_baseline=_restores_a_baseline(self.config))
+            self._benchmark_identity.bind_runtime(self.environment.image_id, required=bool(self.config.benchmark.version))
             if reference_mode:
                 self._run_reference_solution_tests(tests)
             else:
@@ -515,6 +527,8 @@ def run(self) -> dict[str, Any]:
             "metadata": {
                 "suite": self.config.run.suite,
                 "mode": "reference_solution" if reference_mode else "model",
+                "run_id": self._run_id,
+                "benchmark": self._benchmark_identity.metadata(self.records),
                 "result_schema_version": RESULT_SCHEMA_VERSION,
                 "model": model_config,
                 "grader": self.config.grader.model_dump(mode="json"),
@@ -696,6 +710,8 @@ def _run_exploit_audit(self, tests: list[ExecutionTest]) -> dict[str, Any]:
             "metadata": {
                 "suite": self.config.run.suite,
                 "mode": "exploit_audit",
+                "run_id": self._run_id,
+                "benchmark": self._benchmark_identity.metadata(self.records) if self._benchmark_identity else None,
                 "result_schema_version": RESULT_SCHEMA_VERSION,
                 "scoring_version": SCORING_VERSION,
                 "grader": self.config.grader.model_dump(mode="json"),
@@ -1079,10 +1095,13 @@ def _write_outputs(self) -> None:
                 "runtime_image_id": self.environment.image_id,
                 "skills": self._skills_metadata(),
                 "variants": [variant.key for variant in self.variants],
+                "benchmark": next(iter(self.results.values()))["benchmark"],
             },
             "models": {
                 name: {
                     "config": result["model_config"],
+                    "run_id": result["run_id"],
+                    "benchmark": result["benchmark"],
                     "base_model": result["base_model"],
                     "variant": result["variant"],
                     "scores": result["scores"],
@@ -1140,10 +1159,14 @@ def run(self) -> dict[str, Any]:
         Returns:
             Dict with model config, aggregate scores, and individual results.
         """
+        self._benchmark_identity = capture_identity(self.config, self.tests)
+        self._benchmark_identity.bind_runtime(self.environment.image_id, required=bool(self.config.benchmark.version))
         self._run_execution_tests(self.tests)
         summary = self.aggregator.finalize()
         return {
             "model_config": self.model_config.model_dump(mode="json"),
+            "run_id": self._run_id,
+            "benchmark": self._benchmark_identity.metadata(self.records),
             "scoring_version": SCORING_VERSION,
             "usage": self.usage_aggregator.summary(),
             "errored_test_ids": errored_test_ids(self.records),
diff --git a/python/wp_bench/records.py b/python/wp_bench/records.py
index 800c445..dea6253 100644
--- a/python/wp_bench/records.py
+++ b/python/wp_bench/records.py
@@ -16,7 +16,8 @@
 #: field were removed (knowledge track removed).
 #: 2.1: added the "variant" block for skill-injection A/B runs.
 #: 2.2: dropped the "difficulty" field from every record.
-RESULT_SCHEMA_VERSION = "2.2"
+# 2.3: benchmark identity and run_id on every streamed result.
+RESULT_SCHEMA_VERSION = "2.3"
 
 
 def _baseline_variant_info() -> dict[str, Any]:
@@ -77,6 +78,8 @@ def _base_record(
     the builders from drifting (the key set is asserted identical in tests).
     """
     return {
+        "run_id": None,
+        "benchmark": None,
         "test_id": test.id,
         "suite": test.suite,
         "type": "execution",
diff --git a/wp-bench.example.yaml b/wp-bench.example.yaml
index 619714d..680413e 100644
--- a/wp-bench.example.yaml
+++ b/wp-bench.example.yaml
@@ -1,11 +1,16 @@
 # WP-Bench Configuration
 # Copy to wp-bench.yaml and customize. API keys loaded from .env file.
 
+# Optional: assert a public release before spending on models. Use the exact
+# model/grader protocol from `wp-bench benchmark show --version v1`.
+# benchmark:
+#   version: v1
+
 dataset:
   source: local          # 'local' or 'huggingface'
-  name: wp-core-v1       # suite name or HF dataset path
-  # suite: wp-core-v1    # HF only: suite rows to load; defaults to run.suite
-                        # Use --suite wp-cli-v1 to select CLI workflows.
+  name: wp-core       # suite name or HF dataset path
+  # suite: wp-core    # HF only: suite rows to load; defaults to run.suite
+                        # Use --suite wp-cli to select CLI workflows.
 
 # Multiple models to benchmark (runs sequentially, outputs comparison table)
 models:
@@ -46,7 +51,7 @@ grader:
 
 # Run configuration
 run:
-  suite: wp-core-v1
+  suite: wp-core
   # limit: 10            # limit tests per model (null = all); limited runs
   #                      # use seeded stratified selection across categories
   # seed: 1337           # selection seed (same seed = same subset)

From 811f363627462f21c57b314678499d2b31f7fecf Mon Sep 17 00:00:00 2001
From: Jason Adams <jason.adams@automattic.com>
Date: Thu, 22 Oct 2026 20:38:04 -0600
Subject: [PATCH 2/7] feat: build leaderboard data from validated release
 exports

---
 .github/workflows/ci.yml                |   6 +-
 benchmark/README.md                     |   5 +
 python/pyproject.toml                   |   1 +
 python/tests/test_benchmark_releases.py |  57 ++++++++
 python/wp_bench/benchmark.py            |   2 +
 web/README.md                           |  39 ++++++
 web/build_results.py                    | 169 ++++++++++++++++++++++++
 7 files changed, 276 insertions(+), 3 deletions(-)
 create mode 100644 web/README.md
 create mode 100644 web/build_results.py

diff --git a/.github/workflows/ci.yml b/.github/workflows/ci.yml
index f815c77..04c62dd 100644
--- a/.github/workflows/ci.yml
+++ b/.github/workflows/ci.yml
@@ -27,13 +27,13 @@ jobs:
           cache: pip
 
       - name: Install harness with dev tooling
-        run: pip install -e ./python[dev]
+        run: pip install -e ./python[dev,leaderboard]
 
       - name: Lint (ruff)
-        run: ruff check python
+        run: ruff check python web
 
       - name: Type check (mypy)
-        run: mypy python
+        run: mypy python web
 
       - name: Benchmark release integrity
         env:
diff --git a/benchmark/README.md b/benchmark/README.md
index 2597810..0fa99ee 100644
--- a/benchmark/README.md
+++ b/benchmark/README.md
@@ -120,6 +120,11 @@ skills variants, errored/partial runs, duplicate repeats and mixed versions fail
 before output is created. A legitimate failed model answer is still a scored
 answer and remains in the denominator.
 
+The shared [website adapter](../web/README.md) turns these validated bundles
+into the Spacefast display data, including category scores and paired generation
+comparisons. Model names/pairs are editorial inputs; versions and categories
+come from the frozen manifest.
+
 The output is a small, stable ingestion contract, not a website theme:
 
 ```text
diff --git a/python/pyproject.toml b/python/pyproject.toml
index 46dad1e..99c502d 100644
--- a/python/pyproject.toml
+++ b/python/pyproject.toml
@@ -29,6 +29,7 @@ dependencies = [
 ]
 
 [project.optional-dependencies]
+leaderboard = ["numpy>=2.0"]
 http = ["uvicorn>=0.30", "fastapi>=0.111"]
 dev = ["pytest>=8.3", "ruff>=0.16,<0.17", "mypy>=1.10", "types-PyYAML>=6.0"]
 
diff --git a/python/tests/test_benchmark_releases.py b/python/tests/test_benchmark_releases.py
index 5a7cdea..47084f8 100644
--- a/python/tests/test_benchmark_releases.py
+++ b/python/tests/test_benchmark_releases.py
@@ -345,3 +345,60 @@ def execute_artifact(self, artifact: Any, spec: Any) -> ExecutionResult:
     assert streamed == payload["results"]
     assert publication_issues(payload, load_release(root=release_root)) == []
     assert streamed[0]["run_id"] == payload["metadata"]["run_id"]
+
+
+def _website_adapter() -> Any:
+    import importlib.util
+
+    path = Path(__file__).resolve().parents[2] / "web/build_results.py"
+    spec = importlib.util.spec_from_file_location("leaderboard_adapter", path)
+    assert spec and spec.loader
+    module = importlib.util.module_from_spec(spec)
+    spec.loader.exec_module(module)
+    return module
+
+
+def test_website_builds_from_the_shared_export_and_release(release_root: Path, tmp_path: Path) -> None:
+    exported = export_results(inputs(release_root, tmp_path), tmp_path / "exports", root=release_root)
+    module = _website_adapter()
+    bundle = read_json(exported)
+    data = module.build_data(bundle, load_release(root=release_root), {"models": {"model-a": {"label": "Model A", "provider": "OpenAI"}}}, "commit-sha")
+    result = data["results"]
+    assert result["benchmark"]["version"] == "v1"
+    assert result["task_count"] == 2
+    assert result["models"][0]["label"] == "Model A"
+    assert result["models"][0]["overall"]["passed"] == 4
+    assert result["models"][0]["overall"]["rate"] == 1
+    assert result["models"][0]["overall"]["ci"] == [1, 1]
+    target = module.write_data(data, tmp_path / "display")
+    original = (target / "task-results.csv").read_bytes()
+    assert module.write_data(data, tmp_path / "display") == target
+    assert (target / "task-results.csv").read_bytes() == original
+    data["answers"][0]["passed"] = False
+    with pytest.raises(ValueError, match="immutable"):
+        module.write_data(data, tmp_path / "display")
+    assert (target / "task-results.csv").read_bytes() == original
+
+
+@pytest.mark.parametrize("change", ["coverage", "category", "version"])
+def test_website_refuses_incompatible_display_inputs(release_root: Path, tmp_path: Path, change: str) -> None:
+    exported = export_results(inputs(release_root, tmp_path), tmp_path / "exports", root=release_root)
+    bundle = read_json(exported)
+    if change == "coverage":
+        bundle["models"][0]["answers"].pop()
+    elif change == "category":
+        bundle["models"][0]["answers"][0]["category"] = "another-category"
+    elif change == "version":
+        bundle["benchmark"]["version"] = "v2"
+    bundle["fingerprint"] = digest({key: value for key, value in bundle.items()
+                                     if key not in {"snapshot_id", "fingerprint", "updated_at"}})
+    with pytest.raises(ValueError):
+        _website_adapter().build_data(bundle, load_release(root=release_root), {}, "commit-sha")
+
+
+def test_pinned_baseline_release_rejects_skills_ab_before_generation(release_root: Path) -> None:
+    cfg = config(pinned=True)
+    cfg.skills.paths = [Path('some-skill.md')]
+    tests = _parse_execution_suite(release_root / 'datasets/suites/wp-core/execution/tasks.json')
+    with pytest.raises(ValueError, match='Skills A/B runs are diagnostic'):
+        capture_identity(cfg, tests, root=release_root)
diff --git a/python/wp_bench/benchmark.py b/python/wp_bench/benchmark.py
index 4aee71b..4f0bf86 100644
--- a/python/wp_bench/benchmark.py
+++ b/python/wp_bench/benchmark.py
@@ -222,6 +222,8 @@ def capture_identity(config: HarnessConfig, tests: list[ExecutionTest], *,
         ):
             issues.append("Loaded tasks differ from the frozen release")
         protocol = manifest["evaluation"]["definition"]["protocol"]
+        if config.skills.paths:
+            issues.append("Skills A/B runs are diagnostic; this release defines the baseline protocol")
         if config.run.skip_static or config.run.skip_runtime or config.run.continue_on_error:
             issues.append("Diagnostic run settings are enabled")
         if config.run.execution_isolation != "reset_per_test":
diff --git a/web/README.md b/web/README.md
new file mode 100644
index 0000000..3ee48f4
--- /dev/null
+++ b/web/README.md
@@ -0,0 +1,39 @@
+# Leaderboard data adapter
+
+The Spacefast frontend consumes display data built from the repository's public
+benchmark releases. It does not assign benchmark versions or own category rules.
+
+First validate complete result JSONs with `wp-bench benchmark export` (see
+[../benchmark/README.md](../benchmark/README.md)). Then build display data:
+
+```sh
+pip install -e './python[leaderboard]'
+PYTHONPATH=python python web/build_results.py \
+  --export public-data/v1/SNAPSHOT.json \
+  --catalog model-catalog.json --source-commit COMMIT_SHA \
+  --output-dir SITE_SOURCE/data
+```
+
+The small editorial catalog supplies only names, provider labels, generation
+labels, and explicit predecessor pairs:
+
+```json
+{"models":{"provider/model-id":{"label":"Model name","provider":"OpenAI","generation":"Current"}},"pairs":[{"label":"Model family","previous":"older-model-id","current":"provider/model-id"}]}
+```
+
+New models without editorial entries still appear under their API ID. Comparison
+claims are made only for explicitly paired, complete models on the same frozen
+benchmark. Broad categories, task membership, protocol and the public version
+come exclusively from the repository manifest. The adapter verifies the export
+fingerprint and complete repeated task coverage before calculating scores.
+
+The generated `YYYY.MM.DD.HASH/` files are immutable display-data snapshots. The
+website's catalog selects the latest snapshot within each **public benchmark
+version**; visitors see WP-Bench v1/v2 and an update date, while snapshot IDs remain
+in downloads. Keep prior snapshots and their benchmark manifest. Never substitute
+an old-version score for a missing row in a new benchmark.
+
+This adapter has no campaign-specific paths, provider credentials, or model API
+calls. Raw completions remain in private run artifacts. Publish larger scored
+answer bundles separately from the frontend source when retention grows; the
+frontend only needs compact summaries and stable links to those downloads.
diff --git a/web/build_results.py b/web/build_results.py
new file mode 100644
index 0000000..2bbefaa
--- /dev/null
+++ b/web/build_results.py
@@ -0,0 +1,169 @@
+"""Adapt validated WP-Bench exports to the public site's display data.
+
+No model calls or campaign-specific paths. Public versions and categories come
+from the repository release; labels and generation pairs are editorial inputs.
+"""
+from __future__ import annotations
+
+import argparse
+import csv
+import io
+import json
+import statistics
+from datetime import datetime
+from pathlib import Path
+from typing import Any
+
+import numpy as np
+from wp_bench.benchmark import digest, load_release, read_json
+
+
+def interval(values: list[float]) -> list[float]:
+    array = np.array(values, dtype=float)
+    rng = np.random.default_rng(1337)
+    samples = array[rng.integers(0, len(array), size=(10000, len(array)))].mean(axis=1)
+    return [float(value) for value in np.quantile(samples, [0.025, 0.975])]
+
+
+def summarize(rows: list[dict[str, Any]]) -> dict[str, Any]:
+    grouped: dict[tuple[str, str], list[float]] = {}
+    for row in rows:
+        grouped.setdefault((row["suite"], row["test_id"]), []).append(float(row["passed"]))
+    latencies = [row["usage"]["latency_ms"] / 1000 for row in rows if row["usage"].get("latency_ms") is not None]
+    costs = [row["usage"]["cost_usd"] for row in rows if row["usage"].get("cost_usd") is not None]
+    passed = sum(row["passed"] for row in rows)
+    return {"passed": passed, "total": len(rows), "rate": passed / len(rows),
+            "ci": interval([statistics.mean(values) for _, values in sorted(grouped.items())]),
+            "median_seconds": statistics.median(latencies) if latencies else 0,
+            "cost_per_100": 100 * statistics.mean(costs) if len(costs) == len(rows) else None}
+
+
+def build_data(bundle: dict[str, Any], manifest: dict[str, Any], catalog: dict[str, Any],
+               source_commit: str) -> dict[str, Any]:
+    """Check identities and coverage before deriving any displayed scores."""
+    content = {key: value for key, value in bundle.items() if key not in {"snapshot_id", "fingerprint", "updated_at"}}
+    if (bundle.get("schema_version") != "1.0" or digest(content) != bundle.get("fingerprint")
+            or bundle.get("benchmark") != {"version": manifest["version"], "fingerprint": manifest["fingerprint"]}):
+        raise ValueError("Export and frozen benchmark identities differ")
+    definition = manifest["evaluation"]["definition"]
+    tasks = {(suite, test_id): task for suite, component in manifest["evaluation"]["components"].items()
+             for test_id, task in component["tasks"].items()}
+    categories = [{**category, "task_count": sum(
+        definition["category_mapping"][suite][task["category"]] == category["id"]
+        for (suite, _), task in tasks.items())} for category in definition["categories"]]
+    categories = [category for category in categories if category["task_count"]]
+    models: list[dict[str, Any]] = []
+    means: dict[str, dict[tuple[str, str], float]] = {}
+    flat: list[dict[str, Any]] = []
+    seen_models = set()
+    for model in bundle["models"]:
+        name, rows = model["id"], model["answers"]
+        if name in seen_models:
+            raise ValueError("Duplicate model identity")
+        seen_models.add(name)
+        grouped: dict[tuple[str, str], list[float]] = {}
+        run_tasks: dict[tuple[str, str], set[str]] = {}
+        seen_rows = set()
+        for row in rows:
+            key = row["suite"], row["test_id"]
+            row_key = (*key, row["run_id"])
+            if row_key in seen_rows or key not in tasks or not isinstance(row["passed"], bool):
+                raise ValueError("Duplicate, unknown, or ungraded task")
+            seen_rows.add(row_key)
+            expected_category = definition["category_mapping"][key[0]][tasks[key]["category"]]
+            if row["category"] != expected_category:
+                raise ValueError("Category differs from the release mapping")
+            grouped.setdefault(key, []).append(float(row["passed"]))
+            run_tasks.setdefault((row["suite"], row["run_id"]), set()).add(row["test_id"])
+        repeats = definition["protocol"]["repeats"]
+        if set(grouped) != set(tasks) or any(len(values) != repeats for values in grouped.values()):
+            raise ValueError("Model lacks complete, equal repeated task coverage")
+        for (suite, _), ids in run_tasks.items():
+            if ids != set(manifest["evaluation"]["components"][suite]["tasks"]):
+                raise ValueError("A repeat contains only a partial suite")
+        meta = catalog.get("models", {}).get(name, {})
+        provider = meta.get("provider") or ("Google" if name.startswith("gemini/") else "Anthropic" if name.startswith("anthropic/") else "OpenAI" if name.startswith(("gpt-", "openai/", "o1", "o3", "o4")) else "Other")
+        models.append({"id": name, "label": meta.get("label", name), "provider": provider,
+                       "generation": meta.get("generation", "Evaluated"), "repeats": repeats,
+                       "overall": summarize(rows), "categories": {
+                           category["id"]: summarize([row for row in rows if row["category"] == category["id"]])
+                           for category in categories if category["task_count"]}})
+        means[name] = {key: statistics.mean(values) for key, values in grouped.items()}
+        flat.extend({"model": name, **row} for row in rows)
+    if not models:
+        raise ValueError("Export has no models")
+    comparisons = []
+    for pair in catalog.get("pairs", []):
+        previous, current = pair["previous"], pair["current"]
+        if previous not in means or current not in means:
+            continue
+        differences = [means[current][key] - means[previous][key] for key in sorted(tasks)]
+        delta, ci = 100 * statistics.mean(differences), [100 * value for value in interval(differences)]
+        comparisons.append({**pair, "delta_pp": delta, "ci_pp": ci,
+                            "meaningful_improvement": delta >= 3 and ci[0] > 0,
+                            "interpretation": "Improvement across these tasks" if ci[0] > 0 else
+                            "Regression across these tasks" if ci[1] < 0 else "Difference is uncertain on this task set"})
+    gains = sum(pair["delta_pp"] > 0 for pair in comparisons)
+    meaningful = sum(pair["meaningful_improvement"] for pair in comparisons)
+    finding = (f"Newer models scored higher in {gains} of {len(comparisons)} matched comparisons. "
+               f"{meaningful} gains were at least 3 percentage points with a positive paired interval. "
+               "Those pairs show meaningful gains on this task set; progress varies across model families.") if meaningful else (
+               "This release does not establish meaningful improvement across model generations. Use the measured category results to choose a model.")
+    date = datetime.fromisoformat(bundle["updated_at"])
+    result = {"data_schema_version": "1.0", "display_date": date.strftime("%B %d, %Y").replace(" 0", " "),
+              "benchmark": bundle["benchmark"], "commit": source_commit,
+              "dataset_sha256": digest({path: value for path, value in manifest["evaluation"]["files"].items() if path.startswith("datasets/")}),
+              "wordpress_version": definition["wordpress_version"], "scoring_version": definition["scoring_version"],
+              "task_count": len(tasks), "core_task_count": len(manifest["evaluation"]["components"].get("wp-core", {}).get("tasks", {})),
+              "cli_task_count": len(manifest["evaluation"]["components"].get("wp-cli", {}).get("tasks", {})),
+              "categories": categories, "models": sorted(models, key=lambda model: (-model["overall"]["rate"], model["label"])),
+              "comparisons": comparisons, "finding": finding,
+              "methodology": {"settings": f"Frozen {manifest['version']} protocol. First response, baseline without skills; provider default reasoning. Generation and grader settings are recorded in the release definition.",
+                              "limitations": "A fixed public task set cannot represent every WordPress project. Training exposure is unknown. Category sizes vary. Latency and cost estimates depend on provider behavior and API concurrency.",
+                              "cost_note": "Estimated costs use the usage recorded by the harness. Actual billing may differ.",
+                              "comparison_note": "Generation deltas use identical tasks and 95% paired bootstrap intervals (10,000 resamples, seed 1337). Intervals are exploratory and not adjusted for multiple comparisons. A meaningful gain is at least 3 percentage points with a positive interval."},
+              "coverage_note": "Strict behavioral pass rate. Ties share a rank. Category scores use only that category’s tasks. Small gaps may not be meaningful.",
+              "audit_files": []}
+    result["data_version"] = date.strftime("%Y.%m.%d.") + digest({"results": result, "export_fingerprint": bundle["fingerprint"]})[:8]
+    return {"results": result, "manifest": {"benchmark": bundle["benchmark"], "export_fingerprint": bundle["fingerprint"],
+                                             "input_sha256": bundle["input_sha256"], "source_commit": source_commit}, "answers": flat}
+
+
+def write_data(data: dict[str, Any], destination: Path) -> Path:
+    """Keep version-scoped display files immutable and refresh the latest alias."""
+    version = data["results"]["data_version"]
+    release = destination / version
+    body = json.dumps(data["results"], indent=2, allow_nan=False) + "\n"
+    manifest = json.dumps({"data_version": version, **data["manifest"]}, indent=2) + "\n"
+    stream = io.StringIO(newline="")
+    writer = csv.DictWriter(stream, fieldnames=["model", "suite", "test_id", "run_id", "category", "passed", "latency_ms", "cost_usd"])
+    writer.writeheader()
+    for answer in data["answers"]:
+        writer.writerow({**{key: answer[key] for key in ["model", "suite", "test_id", "run_id", "category", "passed"]},
+                         "latency_ms": answer["usage"].get("latency_ms"), "cost_usd": answer["usage"].get("cost_usd")})
+    files = {"results.json": body, "manifest.json": manifest, "task-results.csv": stream.getvalue()}
+    for name, content in files.items():
+        path = release / name
+        if path.exists() and path.read_bytes() != content.encode():
+            raise ValueError("An immutable display-data release would be overwritten")
+    release.mkdir(parents=True, exist_ok=True)
+    for name, content in files.items():
+        (release / name).write_bytes(content.encode())
+    (destination / "results.json").write_text(body)
+    return release
+
+
+def main() -> None:
+    parser = argparse.ArgumentParser(description=__doc__)
+    parser.add_argument("--export", type=Path, required=True)
+    parser.add_argument("--catalog", type=Path, required=True, help="Editorial model labels and generation pairs")
+    parser.add_argument("--output-dir", type=Path, required=True)
+    parser.add_argument("--source-commit", required=True)
+    args = parser.parse_args()
+    bundle = read_json(args.export)
+    data = build_data(bundle, load_release(bundle["benchmark"]["version"]), read_json(args.catalog), args.source_commit)
+    print(json.dumps({"path": str(write_data(data, args.output_dir)), "benchmark": bundle["benchmark"]}))
+
+
+if __name__ == "__main__":
+    main()

From 807d0df64946fdef363c7281159dc2b764140974 Mon Sep 17 00:00:00 2001
From: Jason Adams <jason.adams@automattic.com>
Date: Thu, 22 Oct 2026 20:42:14 -0600
Subject: [PATCH 3/7] fix: keep unavailable response latency explicit

---
 python/tests/test_benchmark_releases.py | 1 +
 web/build_results.py                    | 2 +-
 2 files changed, 2 insertions(+), 1 deletion(-)

diff --git a/python/tests/test_benchmark_releases.py b/python/tests/test_benchmark_releases.py
index 47084f8..7b06812 100644
--- a/python/tests/test_benchmark_releases.py
+++ b/python/tests/test_benchmark_releases.py
@@ -370,6 +370,7 @@ def test_website_builds_from_the_shared_export_and_release(release_root: Path, t
     assert result["models"][0]["overall"]["passed"] == 4
     assert result["models"][0]["overall"]["rate"] == 1
     assert result["models"][0]["overall"]["ci"] == [1, 1]
+    assert result["models"][0]["overall"]["median_seconds"] is None
     target = module.write_data(data, tmp_path / "display")
     original = (target / "task-results.csv").read_bytes()
     assert module.write_data(data, tmp_path / "display") == target
diff --git a/web/build_results.py b/web/build_results.py
index 2bbefaa..d10a1ff 100644
--- a/web/build_results.py
+++ b/web/build_results.py
@@ -34,7 +34,7 @@ def summarize(rows: list[dict[str, Any]]) -> dict[str, Any]:
     passed = sum(row["passed"] for row in rows)
     return {"passed": passed, "total": len(rows), "rate": passed / len(rows),
             "ci": interval([statistics.mean(values) for _, values in sorted(grouped.items())]),
-            "median_seconds": statistics.median(latencies) if latencies else 0,
+            "median_seconds": statistics.median(latencies) if latencies else None,
             "cost_per_100": 100 * statistics.mean(costs) if len(costs) == len(rows) else None}
 
 

From ed6dcfb0d6adb3731700f3c80fb7802321a12879 Mon Sep 17 00:00:00 2001
From: Jason Adams <jason.adams@automattic.com>
Date: Thu, 22 Oct 2026 20:46:39 -0600
Subject: [PATCH 4/7] fix: describe exported cost estimates accurately

---
 web/README.md        | 4 ++++
 web/build_results.py | 2 +-
 2 files changed, 5 insertions(+), 1 deletion(-)

diff --git a/web/README.md b/web/README.md
index 3ee48f4..cd4d410 100644
--- a/web/README.md
+++ b/web/README.md
@@ -37,3 +37,7 @@ This adapter has no campaign-specific paths, provider credentials, or model API
 calls. Raw completions remain in private run artifacts. Publish larger scored
 answer bundles separately from the frontend source when retention grows; the
 frontend only needs compact summaries and stable links to those downloads.
+
+Future display data identifies cost estimates as harness-recorded usage; it does
+not silently claim those values were normalized to uncached price tables. Missing
+latency/cost stays unavailable rather than appearing as zero.
diff --git a/web/build_results.py b/web/build_results.py
index d10a1ff..37941f6 100644
--- a/web/build_results.py
+++ b/web/build_results.py
@@ -110,7 +110,7 @@ def build_data(bundle: dict[str, Any], manifest: dict[str, Any], catalog: dict[s
                "Those pairs show meaningful gains on this task set; progress varies across model families.") if meaningful else (
                "This release does not establish meaningful improvement across model generations. Use the measured category results to choose a model.")
     date = datetime.fromisoformat(bundle["updated_at"])
-    result = {"data_schema_version": "1.0", "display_date": date.strftime("%B %d, %Y").replace(" 0", " "),
+    result = {"data_schema_version": "1.0", "cost_estimate_basis": "recorded_usage", "display_date": date.strftime("%B %d, %Y").replace(" 0", " "),
               "benchmark": bundle["benchmark"], "commit": source_commit,
               "dataset_sha256": digest({path: value for path, value in manifest["evaluation"]["files"].items() if path.startswith("datasets/")}),
               "wordpress_version": definition["wordpress_version"], "scoring_version": definition["scoring_version"],

From c3e901b52655e61d671047096420909778a2b96f Mon Sep 17 00:00:00 2001
From: Jason Adams <jason.adams@automattic.com>
Date: Thu, 22 Oct 2026 20:50:48 -0600
Subject: [PATCH 5/7] fix: type the release catalog explicitly for CI

---
 python/wp_bench/benchmark.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/python/wp_bench/benchmark.py b/python/wp_bench/benchmark.py
index 4f0bf86..279773a 100644
--- a/python/wp_bench/benchmark.py
+++ b/python/wp_bench/benchmark.py
@@ -144,7 +144,7 @@ def freeze_release(version: str, root: Path = PROJECT_ROOT) -> dict[str, Any]:
     if path.exists():
         raise ValueError(f"Benchmark {version} already exists; published releases are immutable")
     catalog_path = root / "benchmark/releases.json"
-    catalog = read_json(catalog_path) if catalog_path.exists() else {
+    catalog: dict[str, Any] = read_json(catalog_path) if catalog_path.exists() else {
         "schema_version": CONTRACT_SCHEMA_VERSION, "releases": [],
     }
     expected = max((int(entry["version"][1:]) for entry in catalog["releases"]), default=0) + 1

From 44c810b8634c3ae48a912f484b857f3ccf513da9 Mon Sep 17 00:00:00 2001
From: Jason Adams <jason.adams@automattic.com>
Date: Thu, 22 Oct 2026 20:52:45 -0600
Subject: [PATCH 6/7] fix: order leaderboard updates by explicit timestamps

---
 web/README.md        | 4 ++++
 web/build_results.py | 7 +++++--
 2 files changed, 9 insertions(+), 2 deletions(-)

diff --git a/web/README.md b/web/README.md
index cd4d410..5c81252 100644
--- a/web/README.md
+++ b/web/README.md
@@ -41,3 +41,7 @@ frontend only needs compact summaries and stable links to those downloads.
 Future display data identifies cost estimates as harness-recorded usage; it does
 not silently claim those values were normalized to uncached price tables. Missing
 latency/cost stays unavailable rather than appearing as zero.
+
+Order result updates by their explicit `updated_at` timestamp, never by the hash
+suffix of a snapshot ID. Select the latest data within each benchmark version;
+updating an archived benchmark does not replace the default newer benchmark.
diff --git a/web/build_results.py b/web/build_results.py
index 37941f6..0ee2c6e 100644
--- a/web/build_results.py
+++ b/web/build_results.py
@@ -10,7 +10,7 @@
 import io
 import json
 import statistics
-from datetime import datetime
+from datetime import datetime, timezone
 from pathlib import Path
 from typing import Any
 
@@ -110,7 +110,10 @@ def build_data(bundle: dict[str, Any], manifest: dict[str, Any], catalog: dict[s
                "Those pairs show meaningful gains on this task set; progress varies across model families.") if meaningful else (
                "This release does not establish meaningful improvement across model generations. Use the measured category results to choose a model.")
     date = datetime.fromisoformat(bundle["updated_at"])
-    result = {"data_schema_version": "1.0", "cost_estimate_basis": "recorded_usage", "display_date": date.strftime("%B %d, %Y").replace(" 0", " "),
+    if date.tzinfo is None:
+        raise ValueError("Export update time must include a timezone")
+    date = date.astimezone(timezone.utc)
+    result = {"data_schema_version": "1.0", "updated_at": date.isoformat(), "cost_estimate_basis": "recorded_usage", "display_date": date.strftime("%B %d, %Y").replace(" 0", " "),
               "benchmark": bundle["benchmark"], "commit": source_commit,
               "dataset_sha256": digest({path: value for path, value in manifest["evaluation"]["files"].items() if path.startswith("datasets/")}),
               "wordpress_version": definition["wordpress_version"], "scoring_version": definition["scoring_version"],

From 1bb190e9f88869870af418d3bd4be07cec6a003a Mon Sep 17 00:00:00 2001
From: Jason Adams <jason.adams@automattic.com>
Date: Thu, 22 Oct 2026 20:58:01 -0600
Subject: [PATCH 7/7] fix: keep repeat exports from reverting current results

---
 python/tests/test_benchmark_releases.py | 2 ++
 python/wp_bench/benchmark_cli.py        | 8 +++++---
 2 files changed, 7 insertions(+), 3 deletions(-)

diff --git a/python/tests/test_benchmark_releases.py b/python/tests/test_benchmark_releases.py
index 7b06812..7eaef80 100644
--- a/python/tests/test_benchmark_releases.py
+++ b/python/tests/test_benchmark_releases.py
@@ -225,6 +225,8 @@ def test_export_preserves_old_snapshots_and_is_idempotent(release_root: Path, tm
     assert first.read_bytes() == original
     catalog = read_json(output / "results.json")
     assert len(catalog["benchmarks"]["v1"]["snapshots"]) == 2
+    assert export_results(paths, output, root=release_root) == first
+    assert read_json(output / "results.json") == catalog
 
 
 def test_export_requires_both_complete_suites_and_independent_repeats(release_root: Path, tmp_path: Path) -> None:
diff --git a/python/wp_bench/benchmark_cli.py b/python/wp_bench/benchmark_cli.py
index 805921d..06e196c 100644
--- a/python/wp_bench/benchmark_cli.py
+++ b/python/wp_bench/benchmark_cli.py
@@ -127,8 +127,9 @@ def export_results(paths: list[Path], output: Path, *, version: str | None = Non
     catalog: dict[str, Any] = read_json(catalog_path) if catalog_path.exists() else {
         "schema_version": CONTRACT_SCHEMA_VERSION, "benchmarks": {},
     }
-    # Same exact inputs are idempotent, including the original update date.
-    if destination.exists():
+    # Same exact inputs preserve both their date and the latest-results pointer.
+    created = not destination.exists()
+    if not created:
         existing = read_json(destination)
         existing_content = {key: value for key, value in existing.items()
                             if key not in {"snapshot_id", "fingerprint", "updated_at"}}
@@ -143,7 +144,8 @@ def export_results(paths: list[Path], output: Path, *, version: str | None = Non
     entry = catalog["benchmarks"].setdefault(manifest["version"], {"current": snapshot_id, "snapshots": []})
     if snapshot_id not in [item["id"] for item in entry["snapshots"]]:
         entry["snapshots"].append({"id": snapshot_id, "path": f"{manifest['version']}/{snapshot_id}.json"})
-    entry["current"] = snapshot_id
+    if created:
+        entry["current"] = snapshot_id
     temporary = catalog_path.with_suffix(".json.tmp")
     temporary.write_text(json.dumps(catalog, indent=2) + "\n")
     temporary.replace(catalog_path)