agentleFS
serionistserionist/claude-session-restore/.claude/skills/review-plan/SKILL.md

review-plan

Skill0 starsChanged 20 days agoMITView on GitHub

Use when a plan in docs/plans/ is drafted and needs checking before any code ("review the plan", "is this plan ready", start-work's review step). A fresh reviewer applies REVIEW.md and lenses.md, refutes each finding and reports only real defects with their scenario, plus improvements; the round is written into the plan's Plan review section; another round only after a blocker or a major.

37 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use implement-plan, review-design and 9 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.
  • JasonPaffSkill

    implement-plan

    Execute implementation plan with structured tracking and validation using subagent architecture

    Also in 5 projects that use this

  • dirkkok101Skill

    review-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

  • SvampenSkill

    review-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

  • PydanticSkill

    review-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-willSkill

    planner

    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

  • sahinSkill

    execute

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

4 rules. 0 are common, 3 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 36 edited versions we read keep these.

  • 3

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Review: Start one fresh agent, one that has not seen this session, with the plan's path, the ground-truth list and lenses.md (this folder).

    line 21
  2. Distinctive:

    No agents in your tool: apply every lens yourself, then refute each finding, and say in the record that the author reviewed their own plan.

    line 27
  3. Distinctive:

    Another round only after a blocker or a major: Commit the edits and go back to step 3: a fresh reviewer reads the edits and the open …

    line 39
  4. Distinctive:

    Every owner call and every improvement has its answer.

    line 48

“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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Review a plan

A plan review catches a design defect while it is still a one-line edit. Not a hunt: a finding is a defect that would really hurt if the plan were followed as written (REVIEW.md, What a finding is), and most are minor.

Steps

  1. The plan, by path: docs/plans/<date>-<slug>.md, never "the latest plan". Commit it first, so the round names a fixed version: git log -1 --format=%h -- <plan>.
  2. The ground truth the reviewer reads: AGENTS.md, REVIEW.md, docs/design.md, the plan in full, its issue (gh issue view <n> --comments), and every file and doc the plan names.
  3. Review. Start one fresh agent, one that has not seen this session, with the plan's path, the ground-truth list and lenses.md (this folder). It reads and nothing else. It applies every lens, refutes each finding by quoting the plan and the rule or code, and returns the confirmed and refuted findings, each with its scenario and an honest severity, the improvements worth making, and a verdict. Don't read the ground truth into this session yourself.
    • Each confirmed blocker or major goes to its own fresh refuter agent, in parallel. One that can't settle it leaves the finding open, never refuted.
    • No agents in your tool: apply every lens yourself, then refute each finding, and say in the record that the author reviewed their own plan.
  4. Write the round into the plan's "Plan review" section as ### Round <n>:
    • How: the reviewer, the model, the date; Verdict:; Now: what has been applied.
    • A table | # | Sev | Finding | Status |: ids P1, P2, ... numbered on across rounds, the finding with its scenario, status open, fixed (what changed), or wontfix: <why>.
    • Refuted findings under "Refuted", each with its reason.
    • Owner calls, when there are any: | # | Question | Recommendation | Answer |, ids Q1, Q2, ...
    • Improvements, each with its answer: taken, left: <why> or later: <issue>.
  5. Act. Edit the plan for each confirmed finding that keeps its design and scope, and mark it fixed. One that changes either goes to the maintainer as a yes or no question with your recommendation, as an owner call. List the improvements to the developer and ask take, leave or later for each.
  6. Another round only after a blocker or a major. Commit the edits and go back to step 3: a fresh reviewer reads the edits and the open findings only. Only minors: fix or record them, and the loop ends.
  7. Verdict: approved when no blocker or major is open, every owner call has its answer, and the plan names its tests and docs and ends with review-code. Otherwise needs-fixes or needs-answers. Commit the plan.

Done means

  • The plan's "Plan review" section says approved, or the maintainer stopped the loop with the open findings recorded.
  • Every owner call and every improvement has its answer.

Never

  • Report a finding without its scenario, or inflate a minor to keep the loop going.
  • Change the plan's design or scope without the maintainer's yes.
  • Approve a plan with an open blocker on a command line, a one-way door or the harness.
  • Drop a finding silently: refuted and wontfix ones stay in the plan.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.