fix(us-275): autoMode must not skip planning Steps 1–3

.agents/skills/ws-activity-report/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: ws-activity-report
-version: 0.3.59
+version: 0.3.60
 description: >-
   Timesheet entries (date, start, end, description) for ws-spec-to-pr /
   ws-spec-to-pr-lite deliveries. Start = earliest bootstrap file creation in

.agents/skills/ws-changelog/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: ws-changelog
-version: 0.3.59
+version: 0.3.60
 description: Summarized task history writer — appends concise, structured task completion records to the project changelog file.
 invocation_names:
   - changelog

.agents/skills/ws-check-harness/SKILL.md

@@ -2,7 +2,7 @@
 name: ws-check-harness
 description: Meta-harness integrity auditor — scans routing, links, portability, integrity digests, instruction duplication, role clarity, and skill composition topology.
 disable-model-invocation: true
-version: 0.3.59
+version: 0.3.60
 invocation_names:
   - check-harness
   - ws-check-harness

.agents/skills/ws-check-workflows/SKILL.md

@@ -1,7 +1,7 @@
 ---
 name: ws-check-workflows
 description: Workflow FSM simulation runner — validates step continuity, state isolation, provider dispatch, and artifact transitions across standard, lite, and multi-spec pipelines.
-version: 0.3.59
+version: 0.3.60
 disable-model-invocation: true
 invocation_names:
   - check-workflows

.agents/skills/ws-classify-complexity/SKILL.md

@@ -1,7 +1,7 @@
 ---
 name: ws-classify-complexity
 description: Pipeline complexity classifier — analyzes a spec against config.json dagThresholds and recommends ws-spec-to-pr-lite or ws-spec-to-pr (standard).
-version: 0.3.59
+version: 0.3.60
 invocation_names:
   - classify-complexity
   - ws-classify-complexity

.agents/skills/ws-cleanup/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: ws-cleanup
-version: 0.3.59
+version: 0.3.60
 disable-model-invocation: true
 description: >-
   Lists disposable workflow leftovers (telemetry, .runtime, fix-pr temps,

.agents/skills/ws-code-review/SKILL.md

@@ -1,7 +1,7 @@
 ---
 name: ws-code-review
 description: Local two-phase code review with fix → re-review loops (max 3). Trigger when reviewing a branch/diff before ship, or when orch Step 6 / lite Step 3 runs.
-version: 0.3.59
+version: 0.3.60
 disable-model-invocation: true
 invocation_names:
   - code-review

.agents/skills/ws-configure-project/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: ws-configure-project
-version: 0.3.59
+version: 0.3.60
 description: Project configuration wizard — detects project settings and interviews config.json sections (including preview.dryRunCommand and optional specMemo).
 invocation_names:
   - configure-project

.agents/skills/ws-doctor/SKILL.md

@@ -1,7 +1,7 @@
 ---
 name: ws-doctor
 description: Workflow skills diagnostic inspector — read-only diagnose of path errors, tool/script recipes, config switches, and missing references across installed ws-* skills.
-version: 0.3.59
+version: 0.3.60
 disable-model-invocation: true
 invocation_names:
   - doctor

.agents/skills/ws-fable-domain/SKILL.md

@@ -1,7 +1,7 @@
 ---
 name: ws-fable-domain
 description: Domain adapter generator & schemas — binds domain authority, minimum evidence sets, observation rules, and fraud definitions for DevOps, Data, Research, and Ops.
-version: 0.3.59
+version: 0.3.60
 invocation_names:
   - ws-fable-domain
   - /ws-fable-domain

.agents/skills/ws-fable-judge/SKILL.md

@@ -1,7 +1,7 @@
 ---
 name: ws-fable-judge
 description: Adversarial audit of claimed work against git diffs and re-run verifications. Trigger after claimed completion, during local review, or before merge.
-version: 0.3.59
+version: 0.3.60
 invocation_names:
   - ws-fable-judge
   - /ws-fable-judge

.agents/skills/ws-fable-method/SKILL.md

@@ -1,7 +1,7 @@
 ---
 name: ws-fable-method
 description: 7-step structured problem-solving methodology — evidence-based problem classification, primary-source investigation, surgical action, and adversarial verification for complex tasks.
-version: 0.3.59
+version: 0.3.60
 invocation_names:
   - ws-fable-method
   - /ws-fable-method

.agents/skills/ws-fix-pr/SKILL.md

@@ -1,7 +1,7 @@
 ---
 name: ws-fix-pr
 description: Single-pass PR thread fixer — resolves active GitHub or ADO PR review threads, applying targeted code fixes and posting progress reports.
-version: 0.3.59
+version: 0.3.60
 disable-model-invocation: true
 invocation_names:
   - fix-pr

.agents/skills/ws-goal-fix-pr/SKILL.md

@@ -1,7 +1,7 @@
 ---
 name: ws-goal-fix-pr
 description: PR thread convergence loop — orchestrates iterative fix-pr rounds until all open PR review threads are resolved and checks pass.
-version: 0.3.59
+version: 0.3.60
 disable-model-invocation: true
 invocation_names:
   - goal-fix-pr

.agents/skills/ws-goal-loop/SKILL.md

@@ -1,7 +1,7 @@
 ---
 name: ws-goal-loop
 description: Generic convergence loop primitive — stateful loop engine providing sentinel management, heartbeat/settle timers, and re-check convergence control.
-version: 0.3.59
+version: 0.3.60
 disable-model-invocation: true
 invocation_names:
   - goal-loop

.agents/skills/ws-implement-tasks/SKILL.md

@@ -1,7 +1,7 @@
 ---
 name: ws-implement-tasks
 description: Task implementation & fix executor — builds planned features following task DAGs or applies surgical defect fixes from code review findings.
-version: 0.3.59
+version: 0.3.60
 disable-model-invocation: true
 invocation_names:
   - implement-tasks

.agents/skills/ws-karpathy-guidelines/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: ws-karpathy-guidelines
-version: 0.3.59
+version: 0.3.60
 description: Micro code diff hygiene guidelines — reduces LLM coding mistakes through surgical changes, minimal diff footprints, and surfacing assumptions.
 license: MIT
 invocation_names:

.agents/skills/ws-plan-interview/SKILL.md

@@ -1,7 +1,7 @@
 ---
 name: ws-plan-interview
 description: Interactive plan interrogation engine — audits implementation plans to uncover hidden assumptions, resolve ambiguities, and refine technical designs.
-version: 0.3.59
+version: 0.3.60
 disable-model-invocation: true
 invocation_names:
   - plan-interview

.agents/skills/ws-plan-to-tasks/SKILL.md

@@ -1,7 +1,7 @@
 ---
 name: ws-plan-to-tasks
 description: Task DAG breakdown generator — transforms approved implementation plans into atomic, dependency-mapped task execution graphs.
-version: 0.3.59
+version: 0.3.60
 disable-model-invocation: true
 invocation_names:
   - plan-to-tasks

.agents/skills/ws-plan-update/SKILL.md

@@ -1,7 +1,7 @@
 ---
 name: ws-plan-update
 description: Post-ship QA delta manager — captures manual QA findings, plans delta fixes, executes modifications, and updates delivery summaries.
-version: 0.3.59
+version: 0.3.60
 disable-model-invocation: true
 invocation_names:
   - plan-update

.agents/skills/ws-plan-verify/SKILL.md

@@ -1,7 +1,7 @@
 ---
 name: ws-plan-verify
 description: Spec compliance scorer (0–10). Pipeline advances only at score ≥ `defaults.minVerifyScore` (default 9); below bar runs scoreAndRefine. Trigger for check-implementation or orch Step 5.
-version: 0.3.59
+version: 0.3.60
 disable-model-invocation: true
 invocation_names:
   - plan-verify

.agents/skills/ws-plan-write/SKILL.md

@@ -1,7 +1,7 @@
 ---
 name: ws-plan-write
 description: Implementation plan generator — transforms feature specifications into structured, technical step-01 implementation plans.
-version: 0.3.59
+version: 0.3.60
 disable-model-invocation: true
 invocation_names:
   - plan-write

.agents/skills/ws-pre-daily/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: ws-pre-daily
-version: 0.3.59
+version: 0.3.60
 disable-model-invocation: true
 description: Standup briefing of the last 36 hours — delivered, made, ongoing, next.
 invocation_names:

.agents/skills/ws-preview/SKILL.md

@@ -1,7 +1,7 @@
 ---
 name: ws-preview
 description: Run the consumer-configured local pipeline review dry-run command without publishing PR threads.
-version: 0.3.59
+version: 0.3.60
 disable-model-invocation: true
 invocation_names:
   - ws-preview

.agents/skills/ws-run-benchmark/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: ws-run-benchmark
-version: 0.3.59
+version: 0.3.60
 description: >-
   Upstream-only package-root fixture compare. Never during ws-spec-to-pr.
   Trigger only on explicit /ws-run-benchmark from the workflow-skills source tree.

.agents/skills/ws-secrets-leak-review/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: ws-secrets-leak-review
-version: 0.3.59
+version: 0.3.60
 description: Secrets & PII leak auditor with an optional runtime-resolving pre-commit hook for local or global installs.
 allowed-tools: Read, Grep, Glob
 invocation_names:

.agents/skills/ws-self-learning/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: ws-self-learning
-version: 0.3.59
+version: 0.3.60
 description: Anti-regression memory engine — consults shared MEMORY before planning/coding and records newly discovered traps into the project knowledge hub.
 invocation_names:
   - self-learning

.agents/skills/ws-senior-developer/SKILL.md

@@ -1,7 +1,7 @@
 ---
 name: ws-senior-developer
 description: Engineering delivery gate for scope control, anti-reinvention, ambiguity stops via user-gate, and pre-ship proof. Invoke via rules.seniorDeveloper or /senior-developer.
-version: 0.3.59
+version: 0.3.60
 invocation_names:
   - senior-developer
   - ws-senior-developer

.agents/skills/ws-shared/gates.md

@@ -34,7 +34,7 @@ Host binding: [`tools.md`](tools.md) § Host-tool binding & dispatch tiers (`ask
 3. If `askQuestionTool` binds `none` → present the **same options** as a short markdown list; wait for user reply. Log: `user-gate-fallback | {gate} | ISO`.
 4. Markdown fallback turn-yielding (mandatory): when a `user-gate` is presented as text/markdown, output ONLY the question and options and MUST NOT emit any tool calls in the same response turn — immediately yield the turn to wait for user input. Emitting a gate plus Step N+1 tool calls in one turn violates this gate.
 5. Cancelled / dismissed → **HS-1** (STOP; re-present; never infer yes).
-6. `autoMode` → zero `user-gate` prompts of any kind (neither modal tool nor markdown) at **every** boundary — entry, transition, G2-code, close, ship, fix-PR; use orch auto-gate table (index 0) to automatically select the recommended option and proceed to the next step without pausing.
+6. `autoMode` → zero `user-gate` prompts of any kind (neither modal tool nor markdown) at **every** boundary — entry, transition, G2-code, close, ship, fix-PR; use orch auto-gate table (index 0) to automatically select the recommended option and proceed to the next step without pausing. **`autoMode` does not skip planning:** the full FSM 0→9 still runs; only gates are automatic. An existing parent feature branch (e.g. `feat/{parent}`) plus a child bug/task slug does not waive Steps 1–3 for the **child slug** — only an explicit user override may shorten planning.
 
 ## Interactive execution cadence (One Step Per Turn)
 

.agents/skills/ws-shared/scripts/workflow_state.cjs

@@ -1171,6 +1171,14 @@ function skippedReason(state, step) {
   return item && item.reason ? String(item.reason) : '';
 }
 
+function stepCompleted(state, step) {
+  return (state.completedSteps || []).map(Number).includes(Number(step));
+}
+
+function stepSkipped(state, step) {
+  return (state.skippedSteps || []).some((row) => Number(row.step) === Number(step));
+}
+
 function requiredAdvanceArtifact(pipeline, next, state) {
   const slug = state.slug || state.us;
   if (!slug) return null;
@@ -1264,6 +1272,29 @@ function validateSnapshot({ stateFile, runFile, indexFile, context, maxStep, pre
     }
     const ledgerFile = path.join(path.dirname(mdPath), 'ac-ledger.json');
     if (next >= 1 && !fs.existsSync(ledgerFile)) errors.push('ac-ledger.json is required before advance');
+    if (Number(next) === 4 && flow === 'standard') {
+      const slug = state.slug || state.us;
+      const usDir = path.dirname(mdPath);
+      const step2Reason = skippedReason(state, 2);
+      if (!stepCompleted(state, 1)) {
+        errors.push('step 1 must be completed before implement');
+      }
+      if (!stepCompleted(state, 2) && step2Reason !== 'interview-not-required') {
+        errors.push('step 2 must be completed or skipped with reason interview-not-required before implement');
+      }
+      const step3Reason = skippedReason(state, 3);
+      if (!stepCompleted(state, 3) && step3Reason !== 'dag-disabled') {
+        errors.push('step 3 must be completed or skipped with reason dag-disabled before implement');
+      }
+      if (slug) {
+        const needRefined = stepCompleted(state, 2) && step2Reason !== 'interview-not-required';
+        const planFile = needRefined ? `step-02-${slug}.plan.refined.md` : `step-01-${slug}.plan.md`;
+        const planPath = path.join(usDir, planFile);
+        if (!fs.existsSync(planPath)) {
+          errors.push(`required artifact missing: ${toRepoRelative(context.repoRoot, planPath, { allowOutside: true })}`);
+        }
+      }
+    }
     if (next >= 6 && fs.existsSync(ledgerFile)) {
       const ledger = JSON.parse(fs.readFileSync(ledgerFile, 'utf8'));
       const boundary = next === 6 ? 'pre-step6' : next >= 9 ? 'ship' : 'step5';
@@ -1384,8 +1415,11 @@ function rebuildIndex(context, config) {
 }
 
 function runValidateCli(config) {
+  let options = {};
   try {
-    const { positional, options } = parseArgs(process.argv.slice(2));
+    const parsed = parseArgs(process.argv.slice(2));
+    options = parsed.options;
+    const { positional } = parsed;
     if (options.help) {
       process.stdout.write('Usage: validate_state.cjs <state|workflowId|rebuild-index> [--pre-advance N] [--repo-root DIR]\n');
       return;
@@ -1409,7 +1443,11 @@ function runValidateCli(config) {
     });
     process.stdout.write(`${JSON.stringify({ ...result, state: toRepoRelative(context.repoRoot, stateFile, { allowOutside: true }) }, null, 2)}\n`);
   } catch (error) {
-    process.stderr.write(`ERROR: ${error.message}\n`);
+    let message = error.message;
+    if (options.preAdvance !== undefined && Number(requirePreAdvanceStep(options.preAdvance)) === 4 && config.pipeline === 'standard' && !message.includes('HS-5')) {
+      message = `${message}; HS-5`;
+    }
+    process.stderr.write(`ERROR: ${message}\n`);
     process.exitCode = 1;
   }
 }

.agents/skills/ws-shared/setup.md

@@ -93,6 +93,7 @@ Standalone `/spec-write` writes `{specsDir}/{slug}.spec.md` only (`plans.specsDi
    | `branch` | `{branch}` |
    | `baseBranch` | `{baseBranch}` |
    ```
+   When `autoMode` is true, append one line immediately after the table: `autoMode: gates automatic; FSM 0→9 intact; no product code until Step 4 after plan artifacts exist on disk.`
    Write this block immediately after flag parsing, before auto-resume. Applies in all modes (normal, auto, dry-run). In `dryRun`, prefix with `[DRY-RUN]`.
 4. **Auto resume** or **Active Resume** (see [Resume / reset](#resume--reset)).
 5. **Identity**: `workflow-id`, `slug`, `us-dir`.
@@ -115,6 +116,8 @@ Standalone `/spec-write` writes `{specsDir}/{slug}.spec.md` only (`plans.specsDi
 
    **`autoMode`:** no `user-gate`. If `{currentBranch}` is `HEAD` (detached): do **not** stay — if `feat/{slug}` exists locally or `ls-remote` shows it, check it out (`checkout-existing`; fetch first when the name is remote-only, per the table below); else create `feat/{slug}` from HEAD (`git checkout -b feat/{slug}`). If `git ls-remote --heads {gitRemote} feat/{slug}` fails for auth/network (non-zero exit, not a missing ref), fall back to **local check only**: create `feat/{slug}` from HEAD only when `git branch --list feat/{slug}` is empty, and log `branch-gate | auto | local-check-only | {branch} | ISO`; never infer "branch absent" from a failed `ls-remote`. Never persist the literal `HEAD` as `state.branch`. Otherwise **stay** on current HEAD (no git mutation), `branchStrategy: stay`. Set `state.branch` = final branch name, `branchStrategy` = `from-current` | `checkout-existing` | `stay`, `baseBranch` = resolved value. Log in `## Gate history`: `branch-gate | auto | stay|from-current|checkout-existing|local-check-only | {branch} | ISO`.
 
+   **Child slug on existing parent branch:** when the workflow slug is a child bug/task id and HEAD is already on a parent feature branch (e.g. `feat/{parent}`), that does not waive Steps 1–3 for the **child slug** — run the full planning chain unless the user explicitly overrides.
+
    **`dryRun`:** prefix `[DRY-RUN]`; show the gate choices (or auto default) and the git commands that **would** run; **no ref mutation** (do not run `git checkout -b`, `git checkout`, or `git fetch`).
 
    **Normal mode — primary `user-gate`** (portable alias `user-gate`; native structured choice when available; markdown fallback; log `user-gate-fallback | feature-branch | ISO` when fallback used). Mark **exactly one** Recommended:

.agents/skills/ws-shared/skill-dependencies.json

@@ -1,6 +1,6 @@
 {
   "version": 1,
-  "packageVersion": "0.3.59",
+  "packageVersion": "0.3.60",
   "upstream": {
     "repo": "jpolvora/workflow-skills",
     "url": "https://github.com/jpolvora/workflow-skills",

.agents/skills/ws-ship-pr/SKILL.md

@@ -1,7 +1,7 @@
 ---
 name: ws-ship-pr
 description: End-to-end PR shipping manager — drives prepare-to-PR checklists, pushes code, creates PRs, waits for CI, and manages convergence.
-version: 0.3.59
+version: 0.3.60
 disable-model-invocation: true
 invocation_names:
   - ship-pr

.agents/skills/ws-show-harness/SKILL.md

@@ -1,7 +1,7 @@
 ---
 name: ws-show-harness
 description: Session harness snapshot generator — reports active skills, rules, precedence hierarchy, and capabilities for the current session.
-version: 0.3.59
+version: 0.3.60
 disable-model-invocation: true
 invocation_names:
   - ws-show-harness

.agents/skills/ws-spec-archive/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: ws-spec-archive
-version: 0.3.59
+version: 0.3.60
 disable-model-invocation: true
 description: >-
   Harvests plansDir delivery facts into specsDir/index.PRD, then proposes

.agents/skills/ws-spec-explain/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: ws-spec-explain
-version: 0.3.59
+version: 0.3.60
 disable-model-invocation: true
 description: >-
   Read-only panorama of a spec or US/issue — status, what it does, what it

.agents/skills/ws-spec-format/SKILL.md

@@ -1,7 +1,7 @@
 ---
 name: ws-spec-format
 description: Specification schema & validator — defines canonical *.spec.md format, section hierarchy, and acceptance criteria rules.
-version: 0.3.59
+version: 0.3.60
 invocation_names:
   - spec-format
   - ws-spec-format

.agents/skills/ws-spec-from-provider/SKILL.md

@@ -1,7 +1,7 @@
 ---
 name: ws-spec-from-provider
 description: Bulk-import open GitHub issues or ADO User Stories into local specs (spec-write + register). Trigger when importing tracker backlog to {specsDir} for ws-spec-list / ws-spec-multi.
-version: 0.3.59
+version: 0.3.60
 disable-model-invocation: true
 invocation_names:
   - spec-from-provider