skills/executing-plans/SKILL.md
Use when executing a written implementation plan in a separate or manual session with batch checkpoints and .planning updates.
npx skillsauth add sipengxie2024/superpower-planning executing-plansInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Load a written plan, review it critically, execute tasks in manageable batches, and keep .planning/ current. This is the manual fallback path when Claude Code dynamic workflows are unavailable or the user wants explicit checkpoints.
Core principle: Follow the plan exactly, verify each batch with evidence, and stop at checkpoints rather than silently drifting.
Announce at start: "I'm using the executing-plans skill to implement this plan in batches."
.planning/plan.md.planning/design.md and .planning/findings.md if they existDefault batch size: first 3 plan tasks, or fewer if tasks are large or tightly coupled.
For each task:
in_progress in .planning/progress.md.planning/findings.md.planning/progress.mdcomplete only when the task's checks have actually passedWhen a batch completes:
.planning/progress.md Task Status Dashboard and session log.planning/findings.md with durable discoveriesIf the user wants a review, invoke superpower-planning:requesting-review with a concrete diff range and the relevant plan task text.
After the checkpoint:
After all tasks complete:
.planning/plan.md.planning/progress.md.planning/progress.mdDo not perform branch merge, PR creation, or destructive cleanup unless the user explicitly asks in the current session.
STOP executing immediately when:
Ask for clarification rather than guessing.
Return to plan review when:
.planning/findings.md.planning/progress.md and .planning/findings.md.planning/progress.md for the full statusmain/master without explicit user consentRelated skills:
superpower-planning:git-worktrees — recommended when starting from a shared or risky branchsuperpower-planning:writing-plans — creates the plan this skill executessuperpower-planning:requesting-review — optional checkpoint or pre-merge reviewsuperpower-planning:receiving-review — process review feedbacksuperpower-planning:archiving — archive the completed planning state when the work is donedevelopment
Use when a spec or requirements exist for a multi-step task and an implementation plan needs to be written before touching code
data-ai
Use when about to claim work is complete, fixed, or passing, before committing or creating PRs.
data-ai
Use when executing implementation plans with parallel task groups or individual tasks too heavy for subagent context limits.
development
Use when implementing any feature or bugfix, before writing implementation code