ln-11-plan-reviewer
Reviews implementation plans against repository evidence and current authoritative guidance. Use before execution to expose gaps and risks; not for completed delivery review.
npx skills add levnikolaevich/claude-code-skills --skill ln-11-plan-reviewer --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
What it does
Performs a read-only, evidence-first second pass over an implementation plan to verify the plan against repository state and current authoritative guidance. The goal is to produce a decision-complete result that is grounded in repository evidence, explicitly states uncertainty, and keeps complexity minimal while preserving safety. The skill follows an ordered checkbox workflow and marks items complete only when action and evidence are complete. It ends with a verdict, decision, and approval applied to incomplete items, prepending a summary line that indicates how many items are complete.
How it works
- It uses a defined execution contract: work through every item in the ordered checklist and mark complete only when its action and required evidence are complete; use N/A or unavailable items as incomplete.
- It routes evidence gathering through multiple tool categories: repository state via native file reads, Git status/history; text and structure reads; symbol and code relationships if needed; and external research only when justified by plan claims.
- It maintains an evidence ledger for material claims and requires checking artifacts, ownership, and policy alignment before applying constraints.
- It ensures a read-only approach to the plan and related repository state, allowing only permitted caches or build artifacts to be generated during the review.
When to use it
Use when you need a formal, evidence-backed review of an implementation plan prior to execution. Triggered when a concrete plan exists and the reviewer seeks to identify gaps, risks, and dependencies grounded in repository evidence.
What it can touch
The skill relies on repository files, manifests, schemas, and executable behavior as evidence; official documentation and standards may be consulted for external claims but are not trusted as local policy. It may reference tool outputs like git status, diff, and history when branch or change context matters. All interactions are read-only regarding the plan.
Caveats
- It adheres to evidence-based principles and does not mutate the source plan.
- If artifacts or external claims are missing, the review may mark findings as incomplete or UNVERIFIED per the guidance in the checklist.
- It requires precise alignment between plan statements and repository evidence; divergent sources are handled by authority preference rules.
# Plan Reviewer **Goal:** Perform a read-only, evidence-first second pass over an implementation plan. Verify the plan; do not execute it. A strong result is decision-complete, grounded in the actual repository, explicit about uncertainty, no more complex than the problem requires, and expressed in the fewest words and execution steps that preserve safety. **Execution contract:** Treat the ordered checkbox workflow below as this skill's Definition of Done. Work through every item in order, and mark it complete only when its action and required evidence are complete. `N/A`, skipped, unavailable, or delegated items remain incomplete. Before returning, apply this skill's verdict, decision, and approval rules to every incomplete item and prepend **Checklist: X/Y complete**<br>**Incomplete: None | section/item — reason; outcome impact; exact next action**; list every incomplete item. ## Tool Routing | Need | Preferred tool | Use it when | Fallback | |---|---|---|---| | Repository instructions and current state | Native file read plus Git | Always read the active host's applicable repository instructions; use `git status`, `diff`, and history when branch or change context matters | Equiv
- Tool Routing
- Evidence Rules
- Checklist
- 1. Establish the Review Contract
- 2. Ground the Plan in the Repository
- 3. Research Corrections and Unknown External Claims
- 4. Review from Every Applicable Perspective
- 5. Run One Independent Challenge Round
- 6. Synthesize a Decision-Complete Result
- Output Contract
What does the ln-11-plan-reviewer skill do?
Reviews implementation plans against repository evidence and current authoritative guidance. Use before execution to expose gaps and risks; not for completed delivery review.
How do I install it?
Run `npx skills add levnikolaevich/claude-code-skills --skill ln-11-plan-reviewer --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From levnikolaevich/claude-code-skills, a repository with 528 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.