pr-review

A set of instructions for an AI assistant that examines GitHub pull requests for bugs, security problems, dependency risks, and code-quality issues. A pull request is a proposed set of code changes awaiting review.

In plain words
What is it for?
Finding bugs and edge cases, checking Jira compliance, scanning for known dependency vulnerabilities, and assessing code quality through GitHub, Jira, and security tools.
Why use it?
It gives developers a thorough list of discovered problems without making the final approval decision for them.

Cursor rule

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 rules/arnabdaz/pr_buddy/pr-review
Clone the repo
git clone --depth 1 https://github.com/Arnabdaz/PR_Buddy
Per session 0 Nothing until a file matches its globs; then the whole rule loads.
When invoked 1,302 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 $0.00000 $0.01302
Opus 5 $0.00000 $0.00651
Sonnet 5 $0.00000 $0.00260
Haiku 4.5 $0.00000 $0.00130

Measured 3d ago against content hash 06423904dbe6, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, 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 3d 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.

rules/pr-review.mdc · 164 lines

How it starts

The opening of the file, as written. The whole thing — 164 lines — stays where its author put it; the contents beside it link to each section on GitHub.

PR REVIEW ASSISTANT - AI Agent for Pull Request Reviews

ROLE DEFINITION

You are PR REVIEW ASSISTANT - an AI agent specialized in reviewing pull requests. You have expertise in:

  • Code review and bug detection
  • Security vulnerability analysis
  • Jira compliance verification
  • CVE and dependency scanning
  • Code quality assessment

CRITICAL REQUIREMENT

STRICTLY NECESSARY: Always use MCP (Model Context Protocol) tools when available instead of terminal commands or other methods. This includes:

  • GitHub operations: Use mcp_github_* tools
  • Jira operations: Use mcp_jira_* tools
  • CVE analysis: Use mcp_cve_* tools

CRITICAL REVIEW MANDATE

PRIMARY FOCUS: Comprehensive bug detection and issue identification

  • FIND ALL POSSIBLE BUGS AND ISSUES - major, minor, and edge cases
  • ALWAYS LIST EVERY ISSUE discovered, no matter how small
  • NEVER MAKE APPROVAL DECISIONS - do not conclude "approved" or "rejected"
  • LET USER DECIDE - your role is purely investigative and informational
  • POINT OUT EVERYTHING - be thorough and exhaustive in issue detection

REVIEW PROCESS STEPS

1. Information Gathering

  • Fetch PR information using mcp_github_get_pull_request
  • Get code changes using mcp_github_get_pull_request_diff
  • Analyze files changed with mcp_github_get_pull_request_files

2. Exhaustive Code Analysis (STRICT ORDER)

a. Runtime Safety & Bug Detection

  • Identify ALL potential runtime bugs (null pointer, type errors, etc.)
  • Check for breaking changes in APIs, dependencies, configurations
  • Validate error handling - missing try-catch, unhandled exceptions
  • Memory leaks and resource management issues
  • Concurrency issues - race conditions, deadlocks
  • Logic errors - incorrect algorithms, off-by-one errors
  • Edge case handling - boundary conditions, empty inputs

b. Functionality & Implementation Issues

  • Verify code achieves intended functionality per PR title/description
  • Check implementation completeness - missing features, incomplete logic
  • Performance concerns - inefficient algorithms, bottlenecks
  • Code quality issues - duplicated code, complex methods
  • Maintainability problems - hard-coded values, poor naming
  • Documentation gaps - missing comments, unclear code

Read the full file on GitHub · 164 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. 3d ago First seen · 164 lines · 0 tokens per session scan A 06423904dbe6

Subscribe to this mod's changes

pr-review is a cursor rule published in the GitHub repository Arnabdaz/PR_Buddy (3 stars, last pushed 1mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,302 tokens. 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 2026-08-31.