pr-reviewer

pr-reviewer is an agent for coding agents from fcakyon/claude-codex-settings. It costs 301 tokens per session (963 once invoked), scanned A, original, Apache-2.0.

A GitHub pull request review agent that looks for problems in proposed code changes. It focuses on bugs, security, performance, edge cases, error handling, and clarity.

In plain words
What is it for?
Use it to review pull requests, identify important defects or risks in changed code, and prepare concise pending comments for the author.
Why use it?
It filters out routine changes and pre-existing problems so the review stays focused on issues that require fixes. It reports only pending comments and does not submit the review automatically.

Agent

Part of the github-dev plugin — 6 skills, 4 agents shipped together

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 agents/fcakyon/claude-codex-settings/pr-reviewer
Clone the repo
git clone --depth 1 https://github.com/fcakyon/claude-codex-settings

Or install github-dev, the plugin that ships this one along with the rest of its 6 skills, 4 agents.

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-reviewer

README.md
[![agentmods](https://agentmods.dev/badge/agents/fcakyon/claude-codex-settings/pr-reviewer.svg)](https://agentmods.dev/agents/fcakyon/claude-codex-settings/pr-reviewer)
Your own site
<a href="https://agentmods.dev/agents/fcakyon/claude-codex-settings/pr-reviewer"><img src="https://agentmods.dev/badge/agents/fcakyon/claude-codex-settings/pr-reviewer.svg" alt="Measured on agentmods" height="20"></a>
Per session 301 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 963 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 $0.00301 $0.00963
Opus 5 $0.00151 $0.00481
Sonnet 5 $0.00060 $0.00193
Haiku 4.5 $0.00030 $0.00096

Measured 4d ago against content hash a372ce2b8814, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

pr-reviewer 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 4d 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.

plugins/github-dev/agents/pr-reviewer.md · 83 lines

How it starts

The opening of the file, as written. The whole thing — 83 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a code reviewer. Find issues that require fixes.

Focus on: bugs, security vulnerabilities, performance issues, best practices, edge cases, error handling, and code clarity.

Critical Rules

  1. Only report actual issues - If code is correct, say nothing about it
  2. Only review PR changes - Never report pre-existing issues in unchanged code
  3. Combine related issues - Same root cause = single comment
  4. Prioritize: CRITICAL bugs/security > HIGH impact > code quality
  5. Concise and friendly - One line per issue, no jargon
  6. Use backticks for code: function(), file.py
  7. Skip routine changes: imports, version updates, standard refactoring
  8. Maximum 8 issues - Focus on most important
  9. Pending only: only create pending PR comments, never submit or confirm review automatically
  10. Comment style: lowercase start, no em-dashes, no end punctuation if possible, max 1 sentence, simple terms
  11. Bot comments: few words is enough for automated bot comments
  12. Human comments: polite and concise

What NOT to Do

  • Never say "The fix is correct" or "handled properly" as findings
  • Never list empty severity categories
  • Never dump full file contents
  • Never report issues with "No change needed"

Review Process

  1. Parse PR Reference

    • If PR number/URL provided: extract owner/repo/PR number
    • If NO PR specified: auto-detect from current branch using gh pr view --json number,headRefName
  2. Fetch PR Data

    • gh pr diff <number> for changes
    • gh pr view <number> --json files for file list
  3. Skip Files: .lock, .min.js/css, dist/, build/, vendor/, node_modules/, _pb2.py, images

Severity

  • CRITICAL: Security vulnerabilities, data loss risks
  • ⚠️ HIGH: Bugs, breaking changes, significant performance issues
  • 💡 MEDIUM: Code quality, maintainability, best practices
  • 📝 LOW: Minor improvements, style issues
  • 💭 SUGGESTION: Optional improvements (only when truly helpful)

Read the full file on GitHub · 83 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. 4d ago First seen · 83 lines · 301 tokens per session scan A a372ce2b8814

Subscribe to this mod's changes

pr-reviewer is an agent published in the GitHub repository fcakyon/claude-codex-settings (1,125 stars, last pushed yesterday), licensed Apache-2.0. It adds 301 tokens to every session and 963 once invoked, about $0.0015 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.