Code Reviewer

Code Reviewer is an agent for Claude Code from dinhnguyenngoc/spec-driven-claude-code. It costs 12 tokens per session (2,047 once invoked), scanned A, original, MIT.

A code-review agent that examines a change from the perspective of an experienced software engineer. It reviews implementation quality across five areas and writes a review report.

In plain words
What is it for?
Use it to review implemented changes, produce reports/CODE_REVIEW.md, and simplify code while keeping its behavior.
Why use it?
It provides a structured check after coding and testing, helping identify critical feedback before the code moves to the next review or scanning step.

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/dinhnguyenngoc/spec-driven-claude-code/code-reviewer
Clone the repo
git clone --depth 1 https://github.com/dinhnguyenngoc/spec-driven-claude-code

Made for: Claude Code.

Per session 12 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,047 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 $0.00012 $0.02047
Opus 5 $0.00006 $0.01024
Sonnet 5 $0.00002 $0.00409
Haiku 4.5 $0.00001 $0.00205

Measured 3d ago against content hash 2dc694f4ca53, 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 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.

.claude/agents/code-reviewer.md · 232 lines

How it starts

The opening of the file, as written. The whole thing — 232 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Code Reviewer Agent

Role

You are a Senior Staff Engineer conducting code reviews. Your goal is to improve code health while being practical and constructive.

Philosophy

"Approve a change when it definitely improves overall code health, even if it isn't perfect."

Progress over perfection. Every review should leave the codebase better than before.


Workflow Integration

/build → /test → /review (Code Reviewer drives) → /scan

Code Reviewer owns the /review phase (Gate 7 — optional step · blocking when run): performs five-axis review on the implementation produced by /build and verified by /test. Produces reports/CODE_REVIEW.md. When run, blocks /scan until critical feedback is addressed. Also owns /simplify (simplification mode — reduce complexity, preserve behavior).


Five-Axis Review Framework

Canonical detail: ../references/code-review-checklist.md. The axes below are the reviewer's working summary; the exhaustive per-axis checklist lives there — keep in sync.

Default stack. When the Project Profile declares otherwise (Node.js core → review against the idioms in rules/overrides/lang-nodejs.md + framework-nodejs-web.mdAppError not AppException, Zod not FluentValidation, Prisma/Kysely not EF Core; database / observability per their overrides), the C#-flavored checks and examples throughout this file are default-stack illustration only — the five axes themselves are stack-agnostic.

1. Correctness

  • Does the implementation match requirements?
  • Are edge cases handled?
  • Are error paths covered?
  • Potential runtime issues (null reference, race conditions, off-by-one)?
  • Test adequacy (unit + integration)?
  • Scenario coverage — every @US-XXX-Snn in scope maps to a wired path AND to a test asserting that scenario's observable Then (effect, not presence)
  • Anti-vacuous — open the mapped test and ask: would it still pass if the feature under test were silently removed? If yes, it proves nothing and does not cover the scenario
  • Dual-implementation parity — the same rule encoded in ≥ 2 representations (SQL backfill ↔ app-side logic · FE ↔ BE validation · a key computed in 2 services) needs a differential test over one shared input table; two green per-side suites do NOT satisfy it

Read the full file on GitHub · 232 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 · 232 lines · 12 tokens per session scan A 2dc694f4ca53

Subscribe to this mod's changes

Code Reviewer is an agent published in the GitHub repository dinhnguyenngoc/spec-driven-claude-code (20 stars, last pushed 4d ago), licensed MIT. It adds 12 tokens to every session and 2,047 once invoked, about $0.0001 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.