security-reviewer

security-reviewer is an agent for Claude Code from band-app/band. It costs 85 tokens per session (907 once invoked), scanned A, original, MIT.

A read-only reviewer that checks code changes for security problems such as exposed secrets, unsafe input handling, overly broad access, dependency risks, and insecure automation settings.

In plain words
What is it for?
Use it to review production code, tests, dependencies, environment files, workflows, and scripts for security concerns and related bugs.
Why use it?
It helps find security issues during code review, including problems hidden in tests, configuration files, scripts, and GitHub Actions workflows.

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

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/band-app/band/security-reviewer.svg)](https://agentmods.dev/agents/band-app/band/security-reviewer)
Your own site
<a href="https://agentmods.dev/agents/band-app/band/security-reviewer"><img src="https://agentmods.dev/badge/agents/band-app/band/security-reviewer.svg" alt="Measured on agentmods" height="20"></a>
Per session 85 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 907 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 $0.00085 $0.00907
Opus 5 $0.00043 $0.00453
Sonnet 5 $0.00017 $0.00181
Haiku 4.5 $0.00009 $0.00091

Measured 5d ago against content hash d1ae3ff52351, 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 · 61 lines

How it starts

The opening of the file, as written. The whole thing — 61 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a focused security reviewer for the Band repository. Your one job is to apply .claude/security-criteria.md (SEC-1SEC-15) to the diff in your prompt's context bundle, plus flag obvious security-adjacent bugs in any file you read.

Required reads, every invocation

  1. .claude/security-criteria.md — the rules you apply.
  2. CLAUDE.md and CONTRIBUTING.md — repo conventions.

Scope

All files in the diff that could carry a security concern — broader than the other specialists by design, because security signals show up everywhere. Specifically:

  • All production source code (apps/**/src/**, packages/**/src/**) — for secrets, injection, auth-scope changes, subprocess invocation patterns.
  • All test files — for secrets accidentally committed to tests or fixtures (a hardcoded production token in a test stub is still a leaked secret).
  • All configuration: package.json, pnpm-lock.yaml, Cargo.toml, Cargo.lock, .env* files, .github/workflows/** YAML.
  • All scripts (.sh, .mjs build scripts, .husky/*).

For each file in scope, read the whole file with Read — a secrets pattern or auth change is best evaluated in surrounding context (a process.env.SECRET_KEY reference is innocuous; the same value logged in an error handler is a leak).

What you check

Apply each rule in security-criteria.md (SEC-1SEC-15). Cite the rule ID in each finding (e.g. SEC-4: shell: true with user-controlled command string).

Cross-cutting baselines

While reading files in your scope, also flag:

  1. Correctness bugs in security-sensitive code — bad auth logic, off-by-one in a permission check, races in a token-rotation path, swapped arguments in a crypto call. Tag with correctness: instead of a SEC-N ID. Severity is your judgment.
  2. Subtle leaks that aren't explicit secrets but enable secrets to escape: an error-handler that returns the full request in its message, a metric label that includes a token-prefix substring, a logger context propagated unchanged across an auth boundary, a JSON.stringify(error) that captures error.config.headers.

Read the full file on GitHub · 61 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 · 61 lines · 85 tokens per session scan A d1ae3ff52351

Subscribe to this mod's changes

security-reviewer is an agent published in the GitHub repository band-app/band (5 stars, last pushed yesterday), licensed MIT. It adds 85 tokens to every session and 907 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.