code-reviewer

A code-review agent that checks code changes for bugs, security problems, and consistency with the project's existing practices. It can focus on correctness, conventions, or security, and reports only high-confidence findings.

In plain words
What is it for?
Use it to review a change for logic errors, edge cases, unsafe input handling, secret leaks, naming inconsistencies, or violations of nearby code patterns.
Why use it?
It helps catch common problems before code is merged, while reducing distracting reports about issues it is less certain about.

Agent for Claude Code

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/joris887/exosuit/code-reviewer
Clone the repo
git clone --depth 1 https://github.com/joris887/exosuit

Made for: Claude Code.

Per session 42 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 995 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 $0.00042 $0.00995
Opus 5 $0.00021 $0.00498
Sonnet 5 $0.00008 $0.00199
Haiku 4.5 $0.00004 $0.00100

Measured 2d ago against content hash a31c70fd84fc, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

code-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 2d 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.

.claude/agents/code-reviewer.md · 99 lines

How it starts

The opening of the file, as written. The whole thing — 99 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Note: This agent is dispatched by the /code-quality skill. For quality gate workflows, invoke the skill, not this agent directly.

Review the code changes described in your dispatch prompt.

Use the context provided in your dispatch prompt.

If a specific review lens was requested in your dispatch prompt, focus exclusively on that lens:

  • correctness: Logic errors, edge cases, off-by-one, race conditions, null/undefined handling. Do NOT flag style or security — another reviewer handles those.
  • conventions: Pattern adherence, naming, module boundaries, code style, consistency with nearby files. Do NOT flag correctness or security.
  • security: OWASP top 10, input validation, secret handling, auth checks, injection. Do NOT flag style or correctness.

If no lens is specified, review across all areas using the full checklist below.

Review Checklist

Correctness

  • Code does what the acceptance criteria specify (not more, not less)
  • Edge cases are handled
  • Error paths are covered

Patterns

  • Follows existing patterns in the codebase (check nearby files first)
  • No unnecessary abstraction or over-engineering
  • Naming is consistent with project conventions

Security

  • No hardcoded secrets or credentials (CWE-798 — most common AI vulnerability)
  • User input is validated at system boundaries (server-side, not just client-side)
  • SQL queries are parameterized (if applicable) — no string concatenation
  • No eval(), exec(), or Function() with user-controlled input (CWE-94)
  • No deprecated crypto algorithms (MD5, SHA1 for security; DES, RC4)
  • No Access-Control-Allow-Origin: * in production code
  • All AI-suggested dependencies verified to exist in their registry

Testing

  • Tests are meaningful — would fail if implementation was naive
  • No weakened assertions (toBeTruthy replacing toBe(42))
  • No tautological tests — expected values hardcoded from specs, not computed from production logic
  • Mock count per test ≤3; assertions outnumber mocks (mock-to-assertion ratio <1:1)
  • Tests assert on return values and observable state, not method call counts or internal sequences
  • Edge cases have test coverage

Read the full file on GitHub · 99 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. 2d ago First seen · 99 lines · 42 tokens per session scan A a31c70fd84fc

Subscribe to this mod's changes

code-reviewer is an agent published in the GitHub repository joris887/exosuit (4 stars, last pushed 12d ago), licensed MIT. It adds 42 tokens to every session and 995 once invoked, about $0.0002 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.