plan-execution
Trigger: execute plan, implement plan, step-by-step execution, plan completion. Execute plans with rollback.
- License 许可
- Apache-2.0
- 最近更新
- 2026-11-10
- Stars
- ★ 0
- Forks
- 0
01 前言 Front Matter
- name
- plan-execution
- triggers
- execute plan, implement plan, step-by-step execution, task execution, plan completion, run plan, do plan
- changelog
- docs/ciclos/cycle28-20260815.md
- description
- Trigger: execute plan, implement plan, step-by-step execution, plan completion. Execute plans with rollback.
- token_budget
- 2850
03 SKILL.md 正文
When to Use
Executing a multi-step plan (from agent, spec, or task list). NOT for 1-file edits (→ quick-executor) or single-concept changes.
WORKFLOW
Step 0: ISOLATE
- Create branch:
git checkout -b plan/<name>(if git available) - Commit baseline state for rollback
Step 1: READ — Load plan completely. Identify tasks, dependencies, success criteria.
Step 2: ORDER — Sequence by dependency. Mark parallelizable items.
Step 3: EXECUTE — One task at a time:
- Implement
- Verify (see gates below)
- If verify fails → rollback:
git checkout -- <modified-files>→ mark BLOCKED → continue unblocked tasks - Mark complete
Step 4: VERIFY — Full test suite + build at end.
Step 5: REPORT — Summary of done/blocked/failed.
FAILURE RULES
- 3 failures total OR 2 consecutive → STOP, report to orchestrator
- Blocked ≠ failed. Blocked tasks remain in queue.
- Never mark done without executing.
OUTPUT
### Plan Execution
| Task | Status | Verified | Notes |
### Summary
- Completed: X/Y
- Blocked: X (reasons)
- Failed: X (reasons)
### Final State
- Branch: [name]
- Last commit: [hash]
Rules
- Isolate before execute. 2. Verify EVERY task. 3. Rollback on failure. 4. 3 total or 2 consecutive failures → STOP.
Anti-Rationalization
| Rationalization | Red Flag | Verification |
|---|---|---|
| "saltar rollback si el paso parece simple" | Ejecutar sin rama ni baseline para revertir | Step 0 ISOLATE: git checkout -b plan/<name> + baseline commit; fallo→git checkout -- <files> y BLOCKED |
| "ejecutar pasos en paralelo aunque el plan sea secuencial" | Paralelizar sin respetar dependencias | Step 2 ORDER: grafo A→B serial vs A∥B paralelo; solo paralelizar si independencia verificada |
| "marcar step done sin evidencia" | Done sin verificar gate del step | Step 3 VERIFY por task: implement→verify gate→rollback si fail→BLOCKED; nunca done sin evidencia |
Red Flags
- Doing work without checking output format → STOP, re-read skill
- Second occurrence of same rationalization → force RED zone
Verification
- Output matches skill ## Output contract + file:line citaton
- cross-ref-check.ps1 → SKILL.md OK
Refs
quick-executor · deep-debugging · quality-gate
docs/skills/plan-execution/reference.md
04 同仓库更多 Skills
web-quality-audit
Comprehensive web audit: performance, a11y, SEO, responsive, animation, design tokens.
ui-engine
UI system — Grid/Flexbox/@layer/:has(), container queries, compositor-only animation, OKLCH tokens, component patterns
baseline-ui
Anti-slop UI — layout, typography, responsive, animation, tokens. Use for cleanup or polish.
skill-creator
Create new AI agent skills following Agent Skills spec -- auto-bootstrap unknown tasks, define structure, register in AGENTS.md
senior-engineer
Apply senior engineering competencies — system design, trade-off analysis, delegation, mentoring, and tech debt strategy
prompt-engineering
Design effective prompts using SPEARS framework — scope, principles, examples, assertions, refinements, with ReAct and multi-agent patterns