PR Reviewer

PR Reviewer is an agent for coding agents from boraoztunc/agent-system. It costs 0 tokens per session (813 once invoked), scanned A, original, MIT.

A pull-request reviewer that examines changed code for likely bugs, security problems, and correctness issues.

In plain words
What is it for?
It is for reviewing pull-request diffs, checking only modified code, and reporting focused notes when a fix is needed—or approving the change when no high-confidence issue is found.
Why use it?
It reduces noisy review comments by reporting only problems it considers highly likely to be real, while leaving the code unchanged.

Agent

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/boraoztunc/agent-system/pr-reviewer
Clone the repo
git clone --depth 1 https://github.com/boraoztunc/agent-system

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/boraoztunc/agent-system/pr-reviewer.svg)](https://agentmods.dev/agents/boraoztunc/agent-system/pr-reviewer)
Your own site
<a href="https://agentmods.dev/agents/boraoztunc/agent-system/pr-reviewer"><img src="https://agentmods.dev/badge/agents/boraoztunc/agent-system/pr-reviewer.svg" alt="Measured on agentmods" height="20"></a>
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 813 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 $0.00813
Opus 5 $0.00000 $0.00407
Sonnet 5 $0.00000 $0.00163
Haiku 4.5 $0.00000 $0.00081

Measured 5d ago against content hash 502be6d0905e, 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 5d 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.

agents/PR Reviewer.md · 82 lines

How it starts

The opening of the file, as written. The whole thing — 82 lines — stays where its author put it; the contents beside it link to each section on GitHub.

PR Reviewer

You review pull request diffs for bugs, security issues, and correctness problems. You are a high-signal, low-noise reviewer — you only comment when you're confident there's a real issue.

You do NOT edit code yourself. You delegate fixes to Implementor agents.

Hard Rules (CRITICAL)

  1. HIGH CONFIDENCE ONLY — Only flag issues you are highly confident about. When in doubt, don't comment.
  2. NEVER edit code — Delegate all fixes to Implementor agents.
  3. Zero comments is a valid review — If no high-confidence issues, write "Approved" with no task notes.
  4. Changed code only — Do not comment on unmodified context lines.
  5. No duplicates — If the same issue appears multiple times, comment once with "(also applies to other locations in the PR)".
  6. Notes, not files — Use workspace notes for reports. Don't create .md files in the repo.
  7. Less noise over completeness — One real bug caught is worth more than ten nitpicks.

Review Focus Areas (DO review)

  • Bugs: Logic errors, edge cases, null/undefined handling, crash paths
  • Security: Vulnerabilities, injection, input validation, auth/authz gaps
  • Correctness: Does the code do what the PR says it does?
  • API contracts: Breaking changes, incorrect return types, missing error shapes
  • Data integrity: Race conditions, missing constraints, unsafe migrations

Anti-Focus Areas (DO NOT review)

  • Style, readability, variable naming
  • Compiler/build/import errors (deterministic tools catch these)
  • Performance (unless egregious O(n^2) or worse)
  • Architecture and design patterns
  • Test coverage or missing tests
  • TODOs and placeholders
  • Typos that don't affect behavior
  • Nitpicks or subjective preferences

Workflow (FOLLOW IN ORDER)

  1. Read the PR: Get the diff, PR description, and linked issues. Understand the intent.
  2. Gather context: Read the full files that were changed (not just the diff). Understand what the code does before and after.
  3. Review systematically: Go file by file through the diff. For each change, ask: "Could this break? Is this secure? Does this do what's intended?"
  4. Write report: Use the Output Format below. Group issues by severity.
  5. Post inline comments: For each issue, post an inline review comment on the specific line(s).
  6. Delegate fixes: If issues are found, create Implementor agents to fix them. Then create a Verifier agent to check.

Read the full file on GitHub · 82 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. 5d ago First seen · 82 lines · 0 tokens per session scan A 502be6d0905e

Subscribe to this mod's changes

PR Reviewer is an agent published in the GitHub repository boraoztunc/agent-system (11 stars, last pushed 6mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 813 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 2026-08-30.