security-auditor

security-auditor is an agent for Claude Code from cdeust/zetetic-team-subagents. It costs 25 tokens per session (11,196 once invoked), scanned A, original, MIT.

A software security reviewer that examines changes and systems for ways attackers could misuse them. It covers areas such as login, encryption, payments, personal data, permissions, dependencies, secrets, and incident response.

In plain words
What is it for?
Use it to model threats, review permissions and cryptography, audit third-party packages, check secret handling, investigate incidents, and assess security-sensitive code changes.
Why use it?
It helps uncover security weaknesses before they cause data loss, unauthorized access, or service disruption. It also checks whether protections work together across the whole attack surface.

Agent for Claude Code

Written for Claude Code: when-to-use in frontmatter. Also seen: model in frontmatter; reads .claude/ paths; mentions subagents.

Part of the zetetic-team-subagents plugin — 16 skills, 2 commands, 23 agents, 6 hooks, 1 MCP server shipped together

Good fit Use it to model threats, review permissions and cryptography, audit third-party packages…

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/cdeust/zetetic-team-subagents/security-auditor
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.

Clone the repo
git clone --depth 1 https://github.com/cdeust/zetetic-team-subagents

Made for: Claude Code.

Or install zetetic-team-subagents, the plugin that ships this one along with the rest of its 16 skills, 2 commands, 23 agents, 6 hooks, 1 MCP server.

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-auditor

README.md
[![agentmods](https://agentmods.dev/badge/agents/cdeust/zetetic-team-subagents/security-auditor.svg)](https://agentmods.dev/agents/cdeust/zetetic-team-subagents/security-auditor)
Your own site
<a href="https://agentmods.dev/agents/cdeust/zetetic-team-subagents/security-auditor"><img src="https://agentmods.dev/badge/agents/cdeust/zetetic-team-subagents/security-auditor.svg" alt="Measured on agentmods" height="20"></a>
Per session 25 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 11,196 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.00025 $0.11196
Opus 5 $0.00013 $0.05598
Sonnet 5 $0.00005 $0.02239
Haiku 4.5 $0.00003 $0.01120

Measured 2d ago against content hash 1f0785231bb0, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade A, and why

security-auditor scanned grade A with 1 finding 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.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

- curl-pipe-to-shell installer: always a refusal; require a package manager or pinned release with verified signature.
agents/security-auditor.md · 450 lines

How it starts

The opening of the file, as written. The whole thing — 450 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are not a personality. You are the procedure. When the procedure conflicts with "ship it, it's probably fine" or "we've never been attacked," the procedure wins.

You adapt to the project's language, deployment surface, and compliance regime. The moves below are stack-agnostic; you apply them using the idioms and tooling of the system you are auditing.

When a change, system, or dependency has a security consequence. Use for threat-model construction, attack-surface enumeration, defense-in-depth review, supply-chain audit, authorization correctness checks, secret-management review, and incident triage. Pair with Dijkstra+Liskov for cryptographic correctness; Lamport for protocol-level interleaving safety; Rejewski for attack-path reverse engineering; Coase for cost-benefit of controls; devops-engineer+Boyd for incident response; engineer for code-level fixes; architect for redesign.

STRIDE threat modeling (Shostack 2014): per-asset decomposition into Spoofing, Tampering, Repudiation, Information disclosure, Denial of service, Elevation of privilege. Unit of analysis is the asset + adversary, not the feature. Source: Shostack, A. (2014). Threat Modeling: Designing for Security. Wiley.

NIST SP 800-63 (current revision): authentication and identity-proofing guidelines; defines IAL/AAL/FAL levels, memorized-secret rules, MFA, session management. Source: NIST SP 800-63-3 (and -A/-B/-C parts), csrc.nist.gov.

SLSA (Supply-chain Levels for Software Artifacts): producer-consumer spec for build integrity (L1–L4: provenance, hermetic builds, reproducibility). Source: slsa.dev, OpenSSF. Sigstore: keyless signing (cosign, rekor, fulcio) for artifacts and images. Source: sigstore.dev, OpenSSF.

Stack-specific idiom mapping:

  • Parameterized queries: SQL driver placeholders ($1, ?, %s with driver binding) — never string interpolation.
  • Secret stores: AWS Secrets Manager, GCP Secret Manager, HashiCorp Vault, K8s Secrets with encryption-at-rest. Not .env in the repo.
  • SBOM: CycloneDX, SPDX. Every production artifact has one.
  • Dependency scanners: pip-audit, npm audit, cargo audit, govulncheck, osv-scanner — detect from lockfiles.
  • Static analyzers: semgrep, bandit, gosec, brakeman — detect from project config.

Read the full file on GitHub · 450 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. 2d ago Changed 1f0785231bb0
  2. 7d ago First seen · 450 lines · 25 tokens per session scan A 5cb62f5d0f1a

Subscribe to this mod's changes

security-auditor is an agent published in the GitHub repository cdeust/zetetic-team-subagents (7 stars, last pushed today), licensed MIT. It adds 25 tokens to every session and 11,196 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.