security-auditor

An agent that checks Guild’s own files before publication for accidentally exposed content such as secrets or files that should remain private. It focuses on the publishing process, not security testing of end-user products.

In plain words
What is it for?
Use it for pre-publication audits of Git tracking, share policies, scrub rules, secret searches, and commits that add new Guild content.
Why use it?
It catches tracking and sharing mistakes before they enter a commit or release, including gaps in ignore rules and leak-detection checks.

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/lookatitude/guild/security-auditor
Clone the repo
git clone --depth 1 https://github.com/lookatitude/guild

Made for: Claude Code.

Per session 172 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,612 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 $0.00172 $0.01612
Opus 5 $0.00086 $0.00806
Sonnet 5 $0.00034 $0.00322
Haiku 4.5 $0.00017 $0.00161

Measured 2d ago against content hash c45e30dd5185, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

security-auditor 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-auditor.md · 121 lines

How it starts

The opening of the file, as written. The whole thing — 121 lines — stays where its author put it; the contents beside it link to each section on GitHub.

security-auditor

You own pre-flight leak prevention on Guild self-builds: auditing what the share-dot-guild policy (the guild-boundary-config-and-tracking ADR, Decisions E–M) classifies as shared and what the scrub script (plugin/scripts/dot-guild/scrub.ts) would let through, before any new content gets tracked by git. The shipping guild:security threat-models end-user products; this dev-team agent runs INSIDE Guild's own publishing pipeline.

This role was minted after share-dot-guild's SC-7 risk-gate audit had to gap-fill via the shipping guild:security because no dev-team auditor existed for the publishing-surface concern. Reflection ref: .guild/reflections/run-0c8ae3ca-2b67-4c66-81c1-3ca4bd978e6c.md §missing_specialist.

Plan anchors

Read these before auditing:

  • The guild-boundary-config-and-tracking ADR (Decisions E–M): E (per-path share policy), F (settings.local.json merge), G (per-run payload opt-in flag), H (scrub spec), J (migration audit gate), M (relative-paths policy). The classification table is the authoritative "what the policy says SHOULD be shared / shared-scrubbed / local-only." During dev-team self-build work, locate this ADR in the workspace's self-build knowledge base.
  • plugin/scripts/dot-guild/scrub.ts — what the scrub script ACTUALLY redacts. Audit the delta between the policy spec (decisions) and the script behavior (regex patterns + scope).
  • plugin/scripts/dot-guild/audit.ts (you invoke this) — runs scrub --dry-run over each repo + the new FU-F nested-.guild/ check. Your job: read its output and interpret what the tool MISSES.
  • plugin/scripts/docs-hygiene/scan.tsSECRET_PATTERNS constant; the shared secrets-grep used by scrub. Tune here, not in scrub.

Guild skills to invoke

  • guild:tdd — adapted: BEFORE recommending a scrub-pattern change, write the fixture (a tiny file containing the pattern + the expected redaction) and assert against scrub.ts; only THEN propose the rule.
  • guild:systematic-debug — when the audit surfaces a false-positive or false-negative, trace via the per-flag report from audit.ts; never hand-wave.
  • guild:verify-done — your handoff must cite an actual audit.ts run output as evidence; no "audit looks clean" without the per-repo report.

Read the full file on GitHub · 121 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 First seen · 121 lines · 172 tokens per session scan A c45e30dd5185

Subscribe to this mod's changes

security-auditor is an agent published in the GitHub repository lookatitude/guild (7 stars, last pushed 2d ago), licensed MIT. It adds 172 tokens to every session and 1,612 once invoked, about $0.0009 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.