Plan Execution
SkillProductivityExecutes an existing implementation plan task by task, preserving its interfaces and constraints, running focused checks after each slice, and stopping when the repository differs from the plan. Use when a written plan or approved checklist already exists. Not for inventing the plan, debugging an unexplained failure, or broad unscoped cleanup.
Available today. Use it from your connected AI after setup.
No other account needed.
Connect ahel once, and every AI you use reads what you have installed.
Then ask your AI: use the Plan Execution skill
What this skill tells your AI
The instructions your AI receives, as published by thiientv/godmode in skills/plan-execution/SKILL.md and read by ahel’s review.
Treat the plan as an executable contract with room for evidence-backed corrections. Do not silently redesign it while implementing.
Before the first task
- Read the plan, referenced design, repository instructions, and current diff.
- Verify the starting branch, worktree, runtime, and baseline checks.
- Check that referenced files, commands, interfaces, and dependencies exist.
- Identify the first task whose result can invalidate the architecture.
Per-task loop
For each task:
- Restate the behavior and files in scope.
- Write or update the failing test/observation before implementation when the plan calls for behavior change.
- Run the focused check and confirm the failure is meaningful.
- Make the smallest change that satisfies the task.
- Run the focused check, then relevant neighboring checks.
- Inspect the diff for scope creep, accidental generated files, and contract drift.
- Record the result, evidence, and remaining limit before moving on.
If a test, dependency, or interface contradicts the plan, stop at a checkpoint and re-shape the affected task. Do not delete a check to make the plan green.
Integration discipline
- Keep each slice buildable when practical.
- Preserve the plan's public names unless an evidence-backed correction is approved.
- Run the broader suite after related slices stabilize.
- Use
completion-verificationfor the final claim.
Completion condition
Execution is complete when every in-scope task has a fresh result, the final diff matches the approved scope, and unverified surfaces are reported.
Signals
- GitHub stars
- 94
- Forks
- 77
- Last commit
- Aug 2026
Advanced
- Catalog kind
- skill
- Gateway key
plan-execution- Source
- github.com/thiientv/godmode