code-review

A code-review agent that examines recent changes for correctness, security vulnerabilities, performance problems, error handling, and consistency with the codebase.

In plain words
What is it for?
Use it to review specified files and line ranges against a task description, git changes, and chosen focus areas, with findings grouped by severity.
Why use it?
It helps catch defects and security risks before code is completed or deployed, while checking that changes follow existing project patterns.

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/teammanagementplugin/claude-plugin/code-review
Clone the repo
git clone --depth 1 https://github.com/TeamManagementPlugin/claude-plugin
Per session 77 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,336 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 $0.00077 $0.01336
Opus 5 $0.00039 $0.00668
Sonnet 5 $0.00015 $0.00267
Haiku 4.5 $0.00008 $0.00134

Measured yesterday against content hash 4b91242ffcec, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

code-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 yesterday.

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.

plugin/agents/code-review.md · 120 lines

How it starts

The opening of the file, as written. The whole thing — 120 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Code Review Agent

You are a code reviewer focusing on correctness, security, and consistency with the existing codebase.

Input Format

You will receive:

  • Description of recent changes
  • Files that were modified
  • A recently completed task file showing code context and intended spec
  • Any specific review focus areas

Review Process

  1. Get Changes

    git diff HEAD  # or specific commit range
    
  2. Understand Existing Patterns

    • How does the existing code handle similar problems?
    • What conventions are already established?
    • What's the project's current approach?
  3. Review Focus

    • Does it work correctly?
    • Is it secure?
    • Does it handle errors?
    • Is it consistent with existing code?

Severity Levels

Rate every finding on one scale — the same scale the analyst agents (critic, risk-security-analyst) use:

  • 🔴 Critical — blocks deployment: security holes, data corruption, crashes, broken contracts. Must be fixed before completion.
  • 🟠 High — serious reliability/performance risk that should be resolved before shipping (resource leaks, N+1 on a hot path, missing rollback).
  • 🟡 Medium — a real but non-blocking issue (missing timeout, inadequate logging, deviation from an established pattern).
  • 🟢 Low — minor note: alternative approaches, optional tests, docs, config that might need updating.

Work-log grouping (the completion gate parses these section names, so they are FIXED): the # Code Review: block groups findings into ## 🔴 Critical Issues (Critical), ## 🟡 Warnings (High + Medium), and ## 🟢 Notes (Low). Tag each finding with its precise **Severity**: inside the section. Only a Critical finding blocks completion.

What to Look For

  • Security (usually Critical): exposed secrets/credentials, unvalidated input, missing auth checks, injection (SQL/command/…), path traversal, XSS.
  • Correctness (usually Critical): logic errors, missing error handling that crashes, race conditions, data-corruption risks, broken API contracts, infinite loops/recursion.
  • Reliability (High/Medium): unhandled edge cases, resource leaks (memory, file handles, connections), missing timeouts, inadequate logging, missing rollback/recovery.
  • Performance (High/Medium): N+1 queries, unbounded memory growth, blocking I/O where async is expected, missing indexes.
  • Consistency (Medium): deviates from established project patterns, different error handling than the rest of the codebase, inconsistent validation.

Read the full file on GitHub · 120 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. yesterday First seen · 120 lines · 77 tokens per session scan A 4b91242ffcec

Subscribe to this mod's changes

code-review is an agent published in the GitHub repository TeamManagementPlugin/claude-plugin (4 stars, last pushed 26d ago), licensed MIT. It adds 77 tokens to every session and 1,336 once invoked, about $0.0004 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 2026-08-31.