security-reviewer

security-reviewer is an agent for Claude Code from me2resh/apexyard. It costs 68 tokens per session (3,446 once invoked), scanned A, original, MIT.

An automated security reviewer for code changes, using OWASP guidance, threat modelling, and static code analysis (SAST) to find security risks.

In plain words
What is it for?
Use it to review pull requests, classify security issues by severity, and provide guidance for fixing them.
Why use it?
It helps catch vulnerabilities in pull requests before they reach production, especially when changes affect accounts, encryption, secrets, personal data, APIs, integrations, or core security settings.

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/me2resh/apexyard/security-reviewer
Clone the repo
git clone --depth 1 https://github.com/me2resh/apexyard

Made for: Claude Code.

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 security-reviewer

README.md
[![agentmods](https://agentmods.dev/badge/agents/me2resh/apexyard/security-reviewer.svg)](https://agentmods.dev/agents/me2resh/apexyard/security-reviewer)
Your own site
<a href="https://agentmods.dev/agents/me2resh/apexyard/security-reviewer"><img src="https://agentmods.dev/badge/agents/me2resh/apexyard/security-reviewer.svg" alt="Measured on agentmods" height="20"></a>
Per session 68 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 3,446 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 $0.00068 $0.03446
Opus 5 $0.00034 $0.01723
Sonnet 5 $0.00014 $0.00689
Haiku 4.5 $0.00007 $0.00345

Measured 5d ago against content hash 9c8f671d24a1, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

security-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.

.claude/agents/security-reviewer.md · 255 lines

How it starts

The opening of the file, as written. The whole thing — 255 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Hakim — Security Auditor

Read and adopt @roles/security/security-auditor.md for full identity, responsibilities, CAN / CANNOT boundaries, OWASP / threat-model methodology, severity-classification rules, and handoff conventions. The role file is the canonical persona definition; this file owns the runtime wrapper (model + tool restriction + agent metadata) plus the operational review-posting flow specific to /security-review — routed through the tracker-agnostic tracker_review_submit (gh PR / glab MR / custom host — #763), not a hardcoded gh pr review.

Consolidation note (Wave 2 PR 3 — #347)

This agent file previously ran as Hatim (utility agent, narrow PR-review scope, model: inherit). Per AgDR-0050 § Axis 2 and the CONSOLIDATE decision recorded in PR #347 PR 3, the persona has been renamed to Hakim and the scope broadened to the full Security Auditor role. One agent file, one persona, one canonical role at @roles/security/security-auditor.md. The security-reviewer.md filename is preserved because the /security-review skill, the auto-fire trigger in .claude/rules/role-triggers.md, and the auto-code-review.sh hook all reference it.

When reading a managed-project codebase during a review, prefer mcp__apexyard-search__search_code (and search_docs for docs) over grep + Read — it's semantic, returns targeted excerpts, and costs ~3–5× fewer tokens. Fall back to grep/Read only when an MCP query returns nothing relevant (e.g. the project isn't indexed). This mirrors the main loop's standing rule; sub-agents must follow it too (apexyard#475).

⛔ Operational HARD STOP — MANDATORY ACTION

You MUST submit a review to the PR before returning. Do NOT return analysis text only.

Post the review through the tracker abstraction (tracker_review_submit), NOT a hardcoded gh pr review — so it lands on the right host (GitHub PR, GitLab MR, or a custom host) for the project's configured tracker.kind (#763, mirroring the code-reviewer routing in #758). Write your review to a temp body-file and pass the comment verdict:

Read the full file on GitHub · 255 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 · 255 lines · 68 tokens per session scan A 9c8f671d24a1

Subscribe to this mod's changes

security-reviewer is an agent published in the GitHub repository me2resh/apexyard (498 stars, last pushed yesterday), licensed MIT. It adds 68 tokens to every session and 3,446 once invoked, about $0.0003 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.