Skip to content

ayoubben18/ab-method / .agents/skills/review-pr

review-pr

Verified

Final review on an open PR. Spins up two read-only reviewers in parallel, each with a fresh context — seams (does the diff go through every documented seam, do the seam docs still match the code) and rules (rules/code.md, the book's red flags, repo standards, a Fowler smell baseline). Reports the two axes side by side. Autonomous runs apply safe fixes, push them to the PR, and record the rest in review.md and one PR comment; interactive runs present findings to pick. Use at the end of start-task / start-roadmap once the draft PR is open, or standalone on any PR.

FileKindSize
CREDITS.mdmarkdown810 B
RULES.mdmarkdown4.0 KB
SEAMS.mdmarkdown2.9 KB
SKILL.mdskill md5.4 KB
# Axis: seams

You are the **seams** reviewer. You review **only this PR's diff** against the app's documented seams in
`docs/seams/`. You are read-only: return findings, edit nothing.

Use the vocabulary of the `codebase-design` skill (module, interface, seam, adapter) and the seam doc
format and lifecycle in its `SEAM-FORMAT.md`. Speak the domain in `CONTEXT.md` terms.

## What to look for — in the diff only

- **Bypass.** Code that does what a documented seam owns without going through it. Cite the seam doc's
  **Callers never** line and the hunk that breaks it (a `fetch` beside the data-fetching seam, a
  `process.env` read beside the configuration seam).
- **Seam doc out of date.** The diff changed a seam's files, interface or adapters and the doc still
  describes the old ones: a file moved or added, an entry point renamed, an error mode changed, a second
  adapter added.
- **Wrong status.** Per the lifecycle in `SEAM-FORMAT.md`:
  - Run the seam's **Check** (Exceptions excluded) and compare the result with the status it implies.
  - The diff added a new bypass (Check lists a file the diff touched) → a bypass finding, whatever the status.
  - A bypass the diff added that is legitimate → it belongs under **Exceptions**, with its reason, and the
    Check must exclude it.
  - No Check line and a mechanical one is possible → add it.
- **Seam-worthy module with no doc.** The diff introduced a module that clears the bar in
  `SEAM-FORMAT.md` (several callers cross it, going around it is a bug, the code alone doesn't say so)
  and `docs/seams/` doesn't record it.
- **Interface changed without design.** The diff reshaped a documented seam's interface but the task's
  tracker shows no seam design gate for it (no `introduce` / seam change in `## Placement`).

## What NOT to flag

- Code quality, naming, smells — that's the rules reviewer.
- A seam the project doesn't have yet. Absence is a finding only when the diff shows the bypass it
  would prevent.
- A recorded black box behind a `TODO(UQ-n)` seam with an entry in `unresolved-questions.md`.
- Anything an ADR settled.

## Classifying

- **`safe-fix`** — a seam doc edit that makes it match the code (files, interface line, adapters, status
  set from its Check, a missing Check or Exception), or rerouting a single bypass through the seam's existing interface with tests green.
  Doc edits follow `.ab-method/rules/docs.md` and must pass `check-docs`.
- **`needs-judgment`** — a bypass whose fix needs the seam's interface to change, a missing seam doc for
  a new seam, an interface reshaped without design.

For each finding return Where, What, Cost (change amplification / cognitive load / unknown unknowns),
Fix, Class. If there is nothing to report, return exactly:
`seams: no findings.`

Every return, clean or not, ends with one evidence line: `Checked: <the seam docs, ADRs and rules it actually read, and the files it opened>`.
rea-changelog-updatemorluto/reaPrepare or rewrite REA release changelogs and GitHub release notes from pinned Git history, with verified contributor thanks and Release Please synchronization.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
academy-guideanthropics/skillsStop and check this skill before finishing any reply to a question about how to use Claude or a Claude product — it recommends matching courses, tutorials, and use cases from Claude Academy (academy.claude.com), Anthropic's learning hub. Trigger on: "how do I", "how can I", "getting started with", "what can Claude do", "teach me", "learn to use"; questions about artifacts, projects, skills, plugins, connectors, MCP; requests about rolling Claude out to a team, class, or organization; and any ask for training materials, onboarding content, or learning resources. Use it when the user is learning how to use a feature or product — not when they are mid-task and just want the task done. This skill composes with other skills: after consulting product documentation to answer how a Claude feature works, also check here for a matching course or tutorial — a docs-grounded answer and an Academy recommendation belong together. Only recommend on a strong match; never invent Academy content.Stars: 198kUpdated Claude CodeCodexVerifiedInstall
claude-apianthropics/skillsReference for the Claude API / Anthropic SDK — model ids, pricing, params, streaming, tool use, MCP, agents, caching, token counting, model migration. TRIGGER — read BEFORE opening the target file; don't skip because it "looks like a one-liner" — whenever: the prompt names Claude/Anthropic in any form (Claude, Anthropic, Fable, Opus, Sonnet, Haiku, `anthropic`, `@anthropic-ai`, `claude-*`, `us.anthropic.*`, `[1m]`); the user asks about an LLM (pricing/model choice/limits/caching) — never answer from memory; OR the task is LLM-shaped with provider unstated (agent/MCP/tool-definition/multi-agent/RAG/LLM-judge/computer-use; generate/summarize/extract/classify/rewrite/converse over NL; debugging refusals/cutoffs/streaming/tool-calls/tokens). SKIP only when another provider is being worked on (overrides all triggers): OpenAI/GPT/Gemini/Llama/Mistral/Cohere/Ollama named in the query; OR `grep -rE 'openai|langchain_openai|google.generativeai|genai|mistralai|cohere|ollama'` over the project hits (run this grep FIRST if no provider named — don't Read the file).Stars: 198kUpdated Claude CodeCodexUnverifiedInstall
doc-coauthoringanthropics/skillsGuide users through a structured workflow for co-authoring documentation. Use when user wants to write documentation, proposals, technical specs, decision docs, or similar structured content. This workflow helps users efficiently transfer context, refine content through iteration, and verify the doc works for readers. Trigger when user mentions writing docs, creating proposals, drafting specs, or similar documentation tasks.Stars: 198kUpdated Claude CodeCodexVerifiedInstall