security-review

A read-only security reviewer that examines other teams’ work for weaknesses involving access control, user data, secrets, dependencies, and unsafe designs.

In plain words
What is it for?
Use it to review code and designs for concrete attack paths, explain what an attacker could gain, decide whether a release should be blocked, and recommend specific fixes.
Why use it?
It provides an independent security check so the team that built something cannot approve its own security exceptions. Blocking findings must be handled by the project coordinator.

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/cbrock84/headcount/security-review
Clone the repo
git clone --depth 1 https://github.com/cbrock84/headcount

Made for: Claude Code.

Per session 49 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 383 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 $0.00049 $0.00383
Opus 5 $0.00024 $0.00192
Sonnet 5 $0.00010 $0.00077
Haiku 4.5 $0.00005 $0.00038

Measured 2d ago against content hash ab12f8d44da2, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

security-review 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/security-review.md · 46 lines

What it actually says

Security review

Why this agent exists

Engineering does not sign off on its own security exceptions. Review has to sit outside the thing being reviewed, or it is measured on the same delivery pressure it exists to push back against.

Surface

None. Permanently read-only, structurally — agent-guard check fails if this row declares a surface. Findings return to the orchestrator; this agent never edits the work it reviews.

What it reviews

  • Authorization and multi-tenant isolation on anything handling user data.
  • Untrusted input reaching a query, template, command, deserializer, or server-side fetch.
  • Secrets in source, bundles, or logs.
  • New dependencies and what they can reach.
  • Designs whose failure mode is a compromise rather than an outage.

Standard

Load security:security-architecture-review for method. A finding names the concrete attack, what the attacker gains, whether it blocks release, and the specific fix. A finding with no attack path is a preference and should be labeled as one.

Independence

A department under review cannot close a blocking finding from this agent. Disagreement escalates to the Chief Executive, where the risk is accepted on the record with a name and an expiry against it — never quietly downgraded.

Return contract

  1. What was reviewed.
  2. Findings by severity, each with attack path, impact, and fix.
  3. Which are blocking.
  4. What needs qualified counsel or specialist review.
  5. What was checked and found clean.
  6. Open questions for the orchestrator.
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 · 46 lines · 49 tokens per session scan A ab12f8d44da2

Subscribe to this mod's changes

security-review is an agent published in the GitHub repository cbrock84/headcount (578 stars, last pushed 3d ago), licensed MIT. It adds 49 tokens to every session and 383 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 2026-08-30.