A multi-pass pull-request review process that looks for real defects, regressions, missing tests, security problems, and broken assumptions, then challenges its own findings.

Use it to review a checked-out branch or a GitHub or GitLab pull request and produce a more reliable list of issues.

A single review can produce false alarms or miss important problems; repeated self-checking helps separate evidence-based issues from weak objections.

pr-reviewer — Skill for Claude Code by mdelapenya

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.

Any agent

npx skills add mdelapenya/coding-skills --skill pr-reviewer

Clone the repo

git clone --depth 1 https://github.com/mdelapenya/coding-skills

Made for: Claude Code.

Or install coding-skills, the plugin that ships this one along with the rest of its 5 skills.

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.

README.md

[![agentmods](https://agentmods.dev/badge/skills/mdelapenya/coding-skills/pr-reviewer/github.svg)](https://agentmods.dev/skills/mdelapenya/coding-skills/pr-reviewer)

Your own site

<a href="https://agentmods.dev/skills/mdelapenya/coding-skills/pr-reviewer"><img src="https://agentmods.dev/badge/skills/mdelapenya/coding-skills/pr-reviewer/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/skills/mdelapenya/coding-skills/pr-reviewer"><img src="https://agentmods.dev/badge/skills/mdelapenya/coding-skills/pr-reviewer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 159 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 4,824 The whole file, excluding the scripts and references it only reads on demand.

Security scan C 1 finding. A grade says what 26 rules found in the file — not that it is safe.

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.

Model	Per session	Once invoked
Fable 5.1	$0.00159	$0.04824
Opus 5	$0.00079	$0.02412
Sonnet 5	$0.00032	$0.00965
Haiku 4.5	$0.00016	$0.00482

Measured 9d ago against content hash 44b0dcaa9fa9, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade C, and why

pr-reviewer scanned grade C with 1 finding 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 9d 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.

Recursive force deletehighDestructive command

rm -rf with a variable or a broad path is one typo away from removing the wrong tree.

The skill never deletes the clone automatically. Cleanup is `rm -rf` on the parent directory (or on the cwd itself, in case 1).

skills/pr-reviewer/SKILL.md · 293 lines

How it starts

The opening of the file, as written. The whole thing — 293 lines — stays where its author put it; the contents beside it link to each section on GitHub.

PR Reviewer

You are the PR's adversarial reviewer. Your job is to find real problems — bugs, regressions, missing tests, broken invariants, security issues, sloppy abstractions — and then immediately challenge yourself so the surviving findings are the ones that genuinely protect the quality of the project.

This skill is built around two ideas:

A single review pass is biased. First-pass adversarial reviews surface plenty of complaints, but many are wrong, irrelevant, or based on misreading the diff. A second self-challenging pass is what turns "noisy nitpicks" into a trustworthy report.
Reviews get stronger with iteration. Each loop refines confidence: false positives drop out, real issues get sharper, and new issues surface as you internalize the change.

Where you can invoke this skill from

Two common ways:

From inside the matching repo (recommended) — pass a bare PR number, or omit $1 to review the currently checked-out PR branch. The skill takes platform / org / repo from git remote -v.
From a barebones directory (no git repo) — pass a PR URL or org/repo#num / group/project!num shorthand. The skill clones the PR into the current directory (when it is empty) or into a temp dir.

The agent (Claude Code, Copilot, Codex, Gemini, …) drives gh / glab (or its equivalent MCP server) for fetch and clone. Each agent's topic file under references/agents/ covers any agent-specific quirks and the loop primitive that agent supports.

Arguments

$1 — PR identifier. Optional. Accepts any of:
- Bare PR number (123) — requires being inside a git repo with the matching remote
- URL — https://github.com/<org>/<repo>/pull/<n> or https://gitlab.com/<group>/<project>/-/merge_requests/<n> (self-managed GitLab hosts work too)
- Shorthand — <org>/<repo>#<n> (GitHub) or <group>/<project>!<n> (GitLab)
- Omitted — review the currently checked-out PR branch (requires being inside a git repo)
--rounds=N — Maximum loop rounds (default: 3). The loop also exits early on convergence.
--no-loop — Run a single round (Steps 1–7) and stop, even if findings have not converged.

Read the full file on GitHub · 293 lines

Files

What ships with it

5 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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.

9d ago First seen · 293 lines · 159 tokens per session scan C 44b0dcaa9fa9

Subscribe to this mod's changes

pr-reviewer is a skill published in the GitHub repository mdelapenya/coding-skills (2 stars, last pushed 3mo ago), licensed MIT. It adds 159 tokens to every session and 4,824 once invoked, about $0.0008 per session on Opus 5. A static security scan graded it C with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Code review

autoreview

Pre-commit/ship code review: Codex default; optional Claude or Pi.

openclaw/clawhub · 20 tokens

rework-rate

Measure and interpret PR rework rate — the emerging 5th DORA metric.

bradygaster/squad · 20 tokens

omh-code-review

This is a Hermes-native code-review workflow skill.

rlaope/oh-my-hermes · 46 tokens

revdiff-plan

Review the last Codex assistant message (plan, analysis, or proposal) with inline annotations in a TUI overlay. Extracts the most recent response from Codex rollout files and opens it in revdiff for review and annotation. Activates on "revdiff-plan", "review plan with revdiff", "annotate plan", "review last response"…

umputun/revdiff · 84 tokens

code-reviewer

Code review specialist focused on patterns, bugs, security, and performance.

RightNow-AI/openfang · 17 tokens

agent-teams-simplify-and-harden

Implementation + audit loop using parallel agent teams with structured simplify, harden, and document passes. Spawns implementation agents to do the work, then audit agents to find complexity, security gaps, and spec deviations, then loops until code compiles cleanly, all tests pass, and auditors find zero issues or…

LeoYeAI/openclaw-master-skills · 115 tokens

pr-reviewer