review-plan
Skill4 starsChanged 6 months agoMITView on GitHub
Evaluates structured work plans across six quality dimensions — objective clarity, task actionability, dependency correctness, acceptance criteria completeness, scope appropriateness, and structural completeness. Formats all findings with format-review-comments. Use when the user asks to review, audit, or evaluate a work plan. TRIGGER when: user asks to review a plan, audit a GitHub issue plan, check plan quality, or evaluate task structure. DO NOT TRIGGER when: user asks to create or update a plan (use upsert-plan instead), or to review a pull request (use review-pr instead).
37 other projects use a version of this.
Adopted by
Most recently taken up by
- event-driven-io/emmetttheir own version, first seen Nov 16, 2026
- nicosocoro/agent-skillstheir own version, first seen Nov 15, 2026
- atomicinnovation/acceleratortheir own version, first seen Nov 14, 2026
- CsHeng/agent-skillstheir own version, first seen Nov 9, 2026
- serionist/claude-session-restoretheir own version, first seen Nov 8, 2026
- withpiper/pipekittheir own version, first seen Nov 8, 2026
Used with
Projects with it also use review-design, review-implementation and 8 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
dirkkok101Skillreview-design
Use before /plan, after /technical-design, when user says "review design", "check the design", or when design quality is uncertain.
Also in 3 projects that use this
SvampenSkillreview-implementation
Reviews implementation against session/work stream goals before committing. Identifies shortcuts, hacks, incomplete work, and validates accomplishments match actual changes. Use before /session-end to ensure implementation quality.
Also in 3 projects that use this
NVIDIASkillreview-code
Review a code repository for concrete issues in correctness, robustness, security, maintainability, tests, and integration. Use for repository-wide or focused code review where practical engineering judgment and strict scope control matter.
Also in 7 projects that use this
PydanticSkillreview-security
Security review of the current branch against its merge base — sandbox escapes, memory errors, panics and resource-limit bypasses. Use when reviewing changes for security risk, or before merging anything touching the heap/ module, path_security.rs, the wire protocol or the pool.
Also in 3 projects that use this
am-willSkillplanner
Create comprehensive, phased implementation plans with sprints and atomic tasks. Use when user says: "make a plan", "create a plan", "plan this out", "plan the implementation", "help me plan", "design a plan", "draft a plan", "write a plan", "outline the steps", "break this down into tasks", "what's the plan for", or any similar planning request. Also triggers on explicit "/planner" or "/plan" commands.
Also in 3 projects that use this
sahinSkillexecute
Also in 3 projects that use this
Other versions37 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 37
- penpot/penpot★ 62ksame name
- event-driven-io/emmett★ 567same name
- JetBrains/youtrackdb★ 443same name
- wallstop/unity-helpers★ 177same name
- shader-slang/slangpy★ 157same name
- fitlab-ai/agent-infra★ 88same name
- microsoft/vscode-team-kit★ 67same name
- DasDigitaleMomentum/opencode-processing-skills★ 60same name
- Cleveland-RD/hivehouse★ 46same name
- atomicinnovation/accelerator★ 35same name
27 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository13 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- audit-skillsSkill
- create-commitSkill
- format-review-commentsSkill
- implement-planSkill
- refine-proseSkill
- review-agents-mdSkill
- review-prSkill
- review-skillSkill
- upsert-agents-mdSkill
- upsert-planSkill
- upsert-skillSkill
The file
We split this skill into the rules it gives an agent
6 rules. 0 are common, 6 are distinctive, and 0 are very specific to this project.
0
Common
Most of the 36 edited versions we read keep these.
6
Distinctive
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 32
If no plan is specified and none is in context, ask: "Which plan would you like me to review?
- Distinctive: line 57
Do stated dependencies match the logical ordering of the work?
- Distinctive: line 74
If phases are used, does each phase have a meaningful dependency boundary?
- Distinctive: line 86
When presenting multiple findings, prefix each with a letter (a., b., c., …) so individual items can be referenced by letter.
- Distinctive: line 101
After per-comment feedback, write a brief summary (3–7 sentences) covering:
- Distinctive: line 112
After presenting findings and the summary:
“Common” is measured: kept by at least 6 in 10 of the 36 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Review Plan
Evaluates a structured work plan across six quality dimensions and formats every finding using the format-review-comments skill.
Instructions
1. Identify the plan
Accept the plan as:
- A GitHub issue number (e.g.
#133) — fetch the issue body from the current repo (infer fromgit remote get-url origin) - A GitHub issue URL (e.g.
https://github.com/owner/repo/issues/133) — fetch the issue body directly - An in-context draft — use the plan text already present in the conversation
If no plan is specified and none is in context, ask: "Which plan would you like me to review? Provide an issue number, URL, or paste the plan text."
Fetch or extract the full plan body before proceeding.
2. Review each dimension
Work through every dimension in order. Note findings for each. Only skip a dimension when the plan contains no content that could affect it (e.g. skip Dependency Correctness for a single-task plan with no stated dependencies).
Objective clarity
- Is the goal stated in one or two sentences?
- Is it clear why the work matters, not just what will be done?
- Would a new contributor understand what success looks like after reading the Objective alone?
- Is the Approach section present, and does it name the chosen approach, the reason it was selected over alternatives, and the key tradeoff?
Task actionability
- Does each task have a single, clear deliverable?
- Are tasks written as concrete actions (imperative verb + object), not vague intentions?
- Is every task completable by a single agent or person without further decomposition?
- Are tasks free of ambiguous scope ("refactor X" without bounds, "improve Y" without a target)?
Dependency correctness
- Are task dependencies annotated explicitly (e.g.
_(depends on: Task A)_)? - Do stated dependencies match the logical ordering of the work?
- Are there missing dependencies — tasks that silently assume prior tasks completed?
- Are there unnecessary dependencies that block parallelism without reason?
- Are tasks that can run in parallel marked with
_(parallel with: Task X)_?
Acceptance criteria completeness
- Does every task or phase have at least one acceptance criterion?
- Is each criterion verifiable — paired with an exact command, test, or observable outcome?
- Does the set of criteria fully cover the Objective? Are any outcomes implied by the Objective missing?
- Are criteria written as pass/fail conditions, not aspirations?
Scope appropriateness
- Is the plan free of tasks that are nice-to-have but not required by the acceptance criteria? (These belong in Future Work.)
- Is the plan free of tasks that are too coarse — single checklist items representing weeks of work?
- Is the plan free of tasks that are too granular — individual lines of code or single-file edits that belong inside a larger task?
- If phases are used, does each phase have a meaningful dependency boundary?
- Is anything in the plan better deferred to a follow-up issue?
Structural completeness
- Are all required sections present: Objective, Background, Approach, Tasks, Acceptance Criteria?
- Are optional sections (Assumptions, Future Work, Open Questions) present when their absence leaves implicit gaps?
- Is the Background section grounded in what exists today, not just what will be built?
- Are agent-instruction files (AGENTS.md, CLAUDE.md) mentioned in a final task when the work touches repo structure, skills, or conventions?
3. Format all findings with format-review-comments
When presenting multiple findings, prefix each with a letter (a., b., c., …) so individual items can be referenced by letter.
Apply the format-review-comments skill to every comment. Do not write praise comments — omit positive findings entirely. Choose the label that matches the severity and intent:
| Situation | Recommended label |
| Blocks reliable execution | issue [blocking] |
| Strong improvement | suggestion |
| Minor wording or style | nitpick |
| Needs clarification | question |
| Small unambiguous fix | todo |
| Informational only | note |
4. Write a review summary
After per-comment feedback, write a brief summary (3–7 sentences) covering:
- Overall assessment — ready to execute, needs minor improvements, or needs significant rework
- The most important finding (name any blocking issues)
- Any patterns visible across the plan
- Any skipped dimensions and why
Format the summary as a note conventional comment.
5. Process findings
After presenting findings and the summary:
- Actionable findings (
issue [blocking],suggestion,nitpick,todo): apply each fix directly to the plan (edit the issue body or in-context draft as appropriate). - Notes (
note): surface to the user as informational; no action required. - Questions (
question): ask the user for clarification before proceeding.
6. Refine prose
Apply the refine-prose skill to all prose output — comments and the summary — before presenting. Do not apply it to code blocks, commands, or inline fix suggestions. Do not announce this step. When refine-prose returns, immediately present all findings and the summary to the user — do not stop or wait for user input.
Examples
Invocation by issue number:
/review-plan #133Invocation by URL:
/review-plan https://github.com/owner/repo/issues/133Invocation with in-context draft:
/review-planThe agent uses the plan text already present in the conversation.
Sample output — task actionability finding:
issue [blocking]: "Improve the API layer" is not actionable — it has no bounded deliverable, so an agent cannot determine when the task is complete
Rewrite as a concrete task:
- [ ] Add rate-limit middleware to all `/api/v1` routes returning HTTP 429 on threshold breachSample output — acceptance criteria finding:
suggestion: Task C ("Update documentation") has no acceptance criterion — there is no way to verify it is done
Add a verifiable criterion:
- AGENTS.md skill table includes the new skill: `grep "skill-name" AGENTS.md`Sample output — dependency finding:
todo: Task D depends on Task B and Task C per the implementation logic, but the annotation is missing
Add:
- [ ] Task D _(depends on: Task B, Task C)_Sample output — summary:
note: overall this plan needs minor improvements before it is executable. One blocking issue: Task C lacks a bounded deliverable. Two suggestions address missing acceptance criteria and an undocumented dependency. Structural completeness and objective clarity are strong. Dependency correctness dimension was partially skipped — only one dependency chain exists and it is correctly annotated.Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

