Simplify ElecTrace around core workflows

.github/workflows/ci.yml

@@ -47,11 +47,6 @@ jobs:
         run: ruff format --check .
       - name: Type check
         run: mypy src
-      - name: Repository hygiene hooks
-        env:
-          SKIP: ruff-check,ruff-format
-        run: pre-commit run --all-files --show-diff-on-failure
-
   test:
     name: Test Python ${{ matrix.python-version }}
     runs-on: ubuntu-latest
@@ -115,6 +110,5 @@ jobs:
           python -m electrace --version
           electrace --version
           electrace --help
-          electrace benchmark --output .ci-benchmark.json
           electrace demo --output .ci-demo
           electrace inspect .ci-demo/electrace-results

.github/workflows/docs.yml

@@ -53,8 +53,3 @@ jobs:
           electrace demo --output .ci-demo
           python -m json.tool .ci-demo/electrace-results/analysis.json > /dev/null
           python -c 'import json; from jsonschema import Draft202012Validator as V; s=json.load(open("schemas/analysis-v0.1.schema.json")); d=json.load(open(".ci-demo/electrace-results/analysis.json")); V.check_schema(s); V(s).validate(d)'
-      - name: Generate and validate synthetic benchmark
-        run: |
-          electrace benchmark --output .ci-benchmark.json
-          python -m json.tool .ci-benchmark.json > /dev/null
-          python -c 'import json; from jsonschema import Draft202012Validator as V; s=json.load(open("schemas/benchmark-v0.1.schema.json")); d=json.load(open(".ci-benchmark.json")); V.check_schema(s); V(s).validate(d)'

.gitignore

@@ -42,9 +42,7 @@ Thumbs.db
 # Local ElecTrace products
 .electrace-demo/
 .electrace-work/
-.ci-benchmark.json
 .ci-demo/
-electrace-benchmark.json
 electrace-results/
 
 # Local logs and scratch

AGENTS.md

@@ -7,7 +7,7 @@ chemical truth.
 
 ## Before changing code
 
-- Read `README.md`, `docs/scientific-scope.md`, and `docs/architecture.md`.
+- Read `README.md`, `docs/workflows.md`, and `docs/scientific-scope.md`.
 - Preserve the row-vector convention `cartesian = fractional @ cell`.
 - Treat frame-local feature identifiers as local only; persistent identity belongs
   to the matching engine.
@@ -28,25 +28,22 @@ mypy src/electrace
 pytest --cov=electrace --cov-report=term-missing
 mkdocs build --strict
 python -m build
-electrace benchmark --output electrace-benchmark.json
 electrace demo --output .electrace-demo
 ```
 
-The deterministic demo and synthetic benchmark are regression fixtures, not
-scientific validation.
+The deterministic demo is a regression fixture, not scientific validation.
 
 ## Layout
 
 - `src/electrace/models.py`: serializable scientific domain models.
 - `src/electrace/geometry.py`: periodic and deformation-aware geometry.
 - `src/electrace/matching.py`: atom/feature assignment and split/merge candidates.
-- `src/electrace/events.py`: event evidence, persistence, and hysteresis.
+- `src/electrace/events.py`: event evidence and consecutive-frame persistence.
 - `src/electrace/io/`: VASP discovery, parsing, descriptor normalization.
 - `src/electrace/plugins/`: capability checks and non-executing command plans.
 - `src/electrace/workflow/`: manifests, staging, explicit execution, and collection.
 - `src/electrace/pipeline.py`: end-to-end analysis and artifact bundle.
 - `src/electrace/report.py`: offline self-contained HTML report.
-- `src/electrace/benchmark.py`: versioned synthetic lineage regression suite.
 - `schemas/`: versioned machine-readable contracts.
 
 ## Scientific change requirements
@@ -55,7 +52,7 @@ Any change to matching costs, thresholds, conservation logic, confidence, or eve
 semantics must include:
 
 1. focused unit tests;
-2. an adversarial or ambiguous case;
+2. an adversarial or boundary case;
 3. an explanation of the scientific assumption;
 4. unchanged or intentionally versioned output schemas;
 5. a changelog entry when behavior visible to users changes.