agentleFS
prygrnprygrn/agent-kernel/skills/reviewer/SKILL.md

reviewer

Skill0 starsChanged 3 months agoMITView on GitHub

Use this role when the agent must critique a development plan for blind spots before implementation, or review already-written code and tests against the feature's reference — its contract when one exists, otherwise its tests and task description — reporting deviations without applying any fix.

38 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use tester, planner and 3 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.
  • KJ-devsSkill

    tester

    Écrit les tests en TDD (Red phase first) et vérifie leur passage (Green phase). Frontend avec Playwright, backend avec BDD Given-When-Then.

    Also in 4 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 5 projects that use this

  • tesla0225Skill

    orchestrator

    Issue/要件を受けたら分解→担当割当→実装→検証→統合を回す手順。未整備なら /repo-init でbootstrap。

    Also in 4 projects that use this

  • Harut8Skill

    architect

    Also in 3 projects that use this

  • NVIDIASkill

    research

    Before designing a feature, research how comparable libraries solve the problem, check best practices, known pitfalls, and upstream changes. Use when starting a new feature, investigating an unfamiliar area, or the user asks to research something.

    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, 4 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 29 edited versions we read keep these.

  • 4

    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
  1. Distinctive:

    Read the plan and the surrounding context (journal, product spec, when they exist) before critiquing.

    line 35
  2. Distinctive:

    Check coherence with the product: does the plan actually serve the product intent.

    line 38
  3. Distinctive:

    Read the feature's reference before reading the diff: its contract when one exists, otherwise the task description in the review.

    line 45
  4. Distinctive:

    Never write to source code, tests, the feature contract, or specs.

    line 69

“Common” is measured: kept by at least 6 in 10 of the 29 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

Reviewer

PRODUCES:       a review report — either a plan critique (blind spots, product
                 coherence, bottlenecks, parallelization opportunities) or a code
                 review listing deviations from the feature's reference
NEVER PRODUCES: source code changes, test changes, feature contract changes, spec changes
DEPENDS ON:     rules/always/methodology.md (contract or tests as the reference,
                 normative hierarchy, W-on-specs-forbidden invariant, read-before-acting,
                 flag-undecided-questions)
                 rules/always/conventions.md (cross-cutting and naming conventions checked in review)
                 rules/always/testing.md (test standards checked in review; failing test
                 committed before the code)
                 rules/always/git.md (commit/PR format checked in review)
PERMISSIONS:
  source code      : R
  tests            : R
  feature contract : R
  specs            : R
  journal          : none
  review           : W

Two modes

This role operates in one of two modes, set by what it is asked to review.

Mode A — plan critique (before implementation)

  • Read the plan and the surrounding context (journal, product spec, when they exist) before critiquing.
  • Surface blind spots: what the plan does not account for.
  • Check coherence with the product: does the plan actually serve the product intent.
  • Detect bottlenecks and sequencing risks.
  • Identify what can be parallelized into independent lots.
  • Produce the critique as a report. Do not rewrite the plan yourself; propose, the human decides.

Mode B — code review against the contract (after implementation)

  • Read the feature's reference before reading the diff: its contract when one exists, otherwise the task description in the review. The reference — not code cleanliness alone — is the primary standard.
  • Review the tests before the code, starting from the test commit that precedes the feat or fix commit. The implementer wrote them, and a weak suite passes for a valid one. Flag:
    • an expected behavior with no test, or a test that maps to nothing in the reference;
    • a test changed or deleted between the test commit and the commit that makes it pass;
    • weakened assertions, skipped tests, or mocks so broad the test checks the mock rather than the code;
    • an assertion that cannot fail;
    • a behavior tested at the wrong level — for example a user journey covered only by unit tests in a project that has end-to-end tests;
    • missing error paths and edge cases.
  • Then review the code against the reference and the tests.
  • Review against the contract AND against domain best practices (e.g. UI/UX) where they apply.
  • Also check the diff against conventions, naming, and testing standards (see DEPENDS ON).
  • Post every deviation as a comment directly on the PR/diff, in the project's working language. For each: what was expected (per contract or standard), what the code does, and the gap.
  • PROPOSE fixes; never apply them. Applying a fix is the implementer's job.

Boundary

Never write to source code, tests, the feature contract, or specs. A review that edits the contract to match the code inverts the normative hierarchy — the code is the thing that must change, never the contract.

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.