pr-review

pr-review is a skill for Claude Code from ymm-oss/fsl. It costs 58 tokens per session (1,927 once invoked), scanned A, original, Apache-2.0.

A review workflow for checking a pull request against its issue, where a pull request is a proposed code change for review.

In plain words
What is it for?
Use it to inspect the issue and diff, run specialized reviews and verification, and return findings ranked by severity without merging.
Why use it?
It independently tests the claimed fix and looks for missing behavior or misleadingly successful checks before a human decides whether to merge.

Skill for Claude Code

Written for Claude Code: installed under .claude/. Also seen: reads .claude/ paths; mentions AGENTS.md.

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/ymm-oss/fsl/pr-review
Any agent
npx skills add ymm-oss/fsl --skill pr-review
Clone the repo
git clone --depth 1 https://github.com/ymm-oss/fsl

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for pr-review

README.md
[![agentmods](https://agentmods.dev/badge/skills/ymm-oss/fsl/pr-review.svg)](https://agentmods.dev/skills/ymm-oss/fsl/pr-review)
Your own site
<a href="https://agentmods.dev/skills/ymm-oss/fsl/pr-review"><img src="https://agentmods.dev/badge/skills/ymm-oss/fsl/pr-review.svg" alt="Measured on agentmods" height="20"></a>
Per session 58 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,927 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00058 $0.01927
Opus 5 $0.00029 $0.00963
Sonnet 5 $0.00012 $0.00385
Haiku 4.5 $0.00006 $0.00193

Measured 6d ago against content hash 4927efe28119, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-05, from the pricing page.

Security

Grade A, and why

pr-review scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/skills/pr-review/SKILL.md · 147 lines

How it starts

The opening of the file, as written. The whole thing — 147 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Review a pull request

Orchestrate an independent review of a pull request, typically one authored by an agent to resolve an issue. This skill is the coordination layer; the per-dimension analysis belongs to the specialized read-only reviewer agents under .claude/agents/. Diagnose and report — do not push fixes onto the branch, and never merge (leave the PR open for human review).

Inputs: a PR number or branch, and the linked issue. If no issue is linked, ask for the intended contract before reviewing.

Lenses that apply to every phase:

  • Trust order per AGENTS.md: contracts and tests, then the Rust implementation, then frozen Python behavior, then prose. The PR description and any implementer conversation are claims, not evidence.
  • Suspect green. A confidently green false negative is more dangerous than a crash, so "the checks pass" is never sufficient on its own.
  • Findings outside the diff still carry a reporting duty: an accepted construct with absent, placeholder, or hollow semantics is a soundness defect wherever it is found.

Phase 0 — Reconstruct the contract

  1. Read the PR (gh pr view, gh pr diff), the linked issue, and any docs/DESIGN-*.md the change touches or should have touched.
  2. Restate, from the issue and repository evidence only: the requested outcome, the affected authority surface, the invariants at risk, and the verification the change needs.
  3. Note which CI lanes actually ran. merge readiness is a bounded fail-fast lane (docs/DESIGN-ci.md), not product verification — record what remains unverified.

Phase 1 — Contract alignment

Map the issue's acceptance criteria to the diff in both directions:

  • Required but missing: criteria with no corresponding change or test.
  • Present but not required: scope creep, speculative fallbacks, compatibility work no contract calls for.

Check authority placement: new behavior lives under rust/, not the frozen src/fslc/; design decisions are backed by an accepted design note; the change is the smallest contract-preserving one.

Read the full file on GitHub · 147 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 6d ago First seen · 147 lines · 58 tokens per session scan A 4927efe28119

Subscribe to this mod's changes

pr-review is a skill published in the GitHub repository ymm-oss/fsl (23 stars, last pushed yesterday), licensed Apache-2.0. It adds 58 tokens to every session and 1,927 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.