/run-execute

name: run-execute
description: Execute work items based on their assigned mode (autopilot, confirm, validate). Supports single-item and multi-item (batch/wide) runs.
version: 1.0.0

<objective> Execute work items based on their assigned mode (autopilot, confirm, validate). Supports both single-item and multi-item (batch/wide) runs. </objective>

<progress_display> Show current position in workflow:

  ### Run Progress
  - [ ] Run initialized (init-run.cjs)
  - [ ] Context loaded
  - [ ] Plan generated  ← {current_step}
  - [ ] Implementation
  - [ ] Tests passing
  - [ ] Code review
  - [ ] Run completed (complete-run.cjs)
  - [ ] Walkthrough generated

Update markers as you progress: [x] = done, ← current = active step </progress_display>

<prerequisites> Before executing scripts, ensure required dependencies are installed:

<step n="1" title="Check yaml Package"> <action>Run: npm list yaml --depth=0 2>/dev/null || echo "NOT_FOUND"</action> <check if="output contains NOT_FOUND"> <output>Installing required dependency: yaml</output> <action>Run: npm install yaml</action> </check> </step>

| Package | Purpose | Install Command | |---------|---------|-----------------| | `yaml` | Parse/stringify state.yaml | `npm install yaml` |

</prerequisites>

<triggers>

• Pending work item ready for execution
• Resumed from interrupted run
• Batch of work items passed from run-plan

</triggers>

<resume_detection critical="true"> Before starting execution, check if resuming an interrupted run:

<step n="0" title="Check for Active Run"> <action>Check state.yaml for runs.active[] array</action>

<check if="runs.active is empty"> <goto step="1">No active run, start fresh</goto> </check>

<check if="runs.active has entries"> <action>Load active run from state.yaml runs.active[0]</action> <action>Read scope (single/batch/wide) and work_items array</action>

<substep n="0a" title="Enumerate Work Item Status"> <action>For EACH work item in runs.active[0].work_items, classify by status:</action> <action>Build status summary from state.yaml (NOT from artifact files):</action> <format> [DONE] {item-id} — completed [WORKING] {item-id} (phase: {current_phase}) — in_progress [PENDING] {item-id} — pending </format> <action>Count: completed={X}, in_progress={Y}, pending={Z}</action> </substep>

<substep n="0b" title="Determine Resume Point for Current Item"> <action>Get current_item from state.yaml</action> <action>Read current_phase from the current item's entry in work_items</action>

<determine_resume_point> Use current_phase from state.yaml to determine resume point:

| current_phase | Resume At | |---------------|-----------| | plan (or unset) | Step 3 (Generate Plan) | | execute | Step 5 (Implementation) | | test | Step 6 (Run Tests) | | review | Step 6b (Code Review) | </determine_resume_point> </substep>

<llm critical="true"> <mandate>NEVER call --complete-item for items with status "completed" — they are already done</mandate> <mandate>NEVER re-execute steps (plan, implement, test) for completed items</mandate> <mandate>ONLY work on the current_item identified in state.yaml</mandate> <mandate>Use current_phase from state.yaml — do NOT infer phase from artifact file existence</mandate> </llm>

<output> Resuming run {run-id} ({scope}) for work item {current_item}. Mode: {mode} Phase: {current_phase} Status: {completed_count} done, {in_progress_count} working, {pending_count} pending Resuming at: Step {step_number} </output> </check> </step> </resume_detection>

<degrees_of_freedom> Varies by mode:

• **Autopilot**: LOW — Execute standard patterns decisively
• **Confirm**: MEDIUM — Present plan, adjust based on feedback
• **Validate**: LOW — Follow approved design precisely

</degrees_of_freedom>

<llm critical="true"> <mandate>USE SCRIPTS — NEVER bypass init-run.cjs or complete-run.cjs</mandate> <mandate>ALWAYS CREATE plan.md — Create plan BEFORE implementation starts (ALL modes)</mandate> <mandate>ALWAYS CREATE test-report.md — Create test report AFTER tests complete</mandate> <mandate>ALWAYS RUN code-review — Invoke code-review skill after tests pass</mandate> <mandate>TRACK ALL FILE OPERATIONS — Every create, modify MUST be recorded</mandate> <mandate>NEVER skip tests — Tests are mandatory, not optional</mandate> <mandate>FOLLOW BROWNFIELD RULES — Read before write, match existing patterns</mandate> </llm>

<artifact_timing critical="true"> Artifacts MUST be created at these points:

| Artifact | When Created | Created By | |----------|--------------|------------| | run.md | Start of run | init-run.cjs script | | plan.md | BEFORE implementation (Step 4) | Agent using template | | test-report.md | AFTER tests pass (Step 6) | Agent using template | | review-report.md | AFTER test report (Step 6b) | code-review skill | | walkthrough.md | After run completes (Step 8) | walkthrough-generate skill |

For batch runs: Append each work item's section to plan.md and test-report.md. </artifact_timing>

<flow> <step n="1" title="Initialize Run"> <hard_gate> ⛔ HARD GATE - SCRIPT EXECUTION REQUIRED ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ You MUST call init-run.cjs script. DO NOT use mkdir or create files directly. The script creates BOTH the folder AND run.md file. If you skip this, state.yaml will be inconsistent. </hard_gate>

<action>Prepare work items JSON array:</action> <code>

For single item:

node scripts/init-run.cjs {rootPath} {workItemId} {intentId} {mode}

For batch/wide (multiple items):

node scr