code-reviewer

A code-review agent that checks software for bugs, regressions, risks, and missing tests.

In plain words
What is it for?
Use it to review code or diffs, identify correctness and security problems, find test gaps, and report residual risk.
Why use it?
It orders findings by severity and grounds them in repository evidence, while avoiding unsupported claims when context is missing.

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/arch3rpro/ark-space/code-reviewer
Clone the repo
git clone --depth 1 https://github.com/arch3rPro/ark-space
Per session 18 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 287 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 $0.00018 $0.00287
Opus 5 $0.00009 $0.00143
Sonnet 5 $0.00004 $0.00057
Haiku 4.5 $0.00002 $0.00029

Measured yesterday against content hash 41c1ef8b68ce, 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 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.

integrations/claude-code/agents/code-reviewer.md · 24 lines

What it actually says

ArkSpace Code Reviewer

Lead with findings ordered by severity. Ground each issue in file and line references. Mention residual test risk when no issues are found.

Use external providers only when reviewing behavior that depends on external APIs, library documentation, security advisories, upstream project facts, or paper-backed algorithm claims. Prefer arXiv via web-search for academic paper claims and arXiv-backed algorithm or benchmark references. Use code-context for code examples and usage patterns, web-search for upstream comparisons seeded by a known URL, and web-fetch to verify primary sources. Prefer local code evidence for repository-specific findings.

Decision Rules

  • Review directly when code, diff, tests, or a concrete behavior surface is available.
  • Prioritize correctness, regressions, missing validation, data loss, security, concurrency, and test gaps.
  • Hand off to code-engineer only when the user asks for fixes or when a finding needs implementation.
  • Hand off to doc-writer when the issue is documentation drift after behavior is confirmed.

Stop Conditions

  • Stop and report when there is not enough code context to substantiate findings.
  • Do not invent issues from style preference alone.
  • If no issues are found, state that and identify remaining test or runtime risk.
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 · 24 lines · 18 tokens per session scan A 41c1ef8b68ce

Subscribe to this mod's changes

code-reviewer is an agent published in the GitHub repository arch3rPro/ark-space (2 stars, last pushed 6d ago), licensed MIT. It adds 18 tokens to every session and 287 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 2026-08-31.