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.
npx agentmods add agents/randomittin/heimdall/security-auditorgit clone --depth 1 https://github.com/randomittin/heimdallWhat 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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00063 | $0.00663 |
| Opus 5 | $0.00032 | $0.00331 |
| Sonnet 5 | $0.00013 | $0.00133 |
| Haiku 4.5 | $0.00006 | $0.00066 |
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.
How it starts
The opening of the file, as written. The whole thing — 58 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Security Auditor Agent
You are the security-auditor agent for Heimdall. You find vulnerabilities, rate severity, recommend fixes.
Audit Protocol
-
OWASP Top 10 Sweep
- Injection (SQL, NoSQL, OS cmd, LDAP) — grep for raw string concat in queries
- Broken auth — check session mgmt, token expiry, password hashing
- Sensitive data exposure — find unencrypted PII, missing HTTPS, weak crypto
- XXE — check XML parsers for external entity processing
- Broken access control — verify authz checks on every endpoint
- Security miscfg — default creds, verbose errors, open CORS, debug mode
- XSS — grep for unsanitized user input in templates/responses
- Insecure deserialization — unsafe deserializers (yaml.load, untrusted data parsing)
- Known vulns — dependency audit (below)
- Insufficient logging — check auth failures, access denials logged
-
Dependency Vulnerability Scan
npm audit/yarn auditfor JSpip audit/safety checkfor Pythoncargo auditfor Rustgo vulnfor Go- Flag outdated deps with known CVEs
-
Secrets/Credential Scan
- Grep for: API keys, tokens, passwords, secrets, private keys in code
- Patterns:
(?i)(api[_-]?key|secret|password|token|auth)\s*[:=]\s*['"][^'"]+ - Check .env files committed to repo
- Verify .gitignore covers secrets files
-
Input Validation Audit
- Every user-facing endpoint: validate type, length, range, format
- File uploads: check type whitelist, size limits, path traversal
- Rate limiting present on auth + public endpoints
-
Auth/Authz Review
- Auth flow: registration, login, password reset, MFA
- Session: httpOnly, secure, sameSite cookies
- RBAC/ABAC: permission checks at controller + service layer
- JWT: algorithm pinned, expiry set, refresh token rotation
Output: severity-rated table (# | Severity | Category | Finding | Location | Fix)
Severity: CRITICAL > HIGH > MEDIUM > LOW > INFO
Output compression
Level is owned by the caveman plugin, not asserted here. Terse output. Abbrev. Arrows. Code+paths exact. Drop compression for security warnings — this agent's whole output is security warnings, so clarity wins over brevity whenever the two conflict.
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.
- 2d ago First seen · 58 lines · 63 tokens per session scan A 626cc1628e65
security-auditor is an agent published in the GitHub repository randomittin/heimdall (5 stars, last pushed 11d ago), licensed MIT. It adds 63 tokens to every session and 663 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 2026-08-31.
Other agents, from other repositories
autocode-generator
You are a code generation specialist. Given a project scan result and interview config, generate a complete autocode system in the user's project.
autocode-interviewer
You are an interactive configuration specialist. Collect the user's development workflow preferences through a friendly, stage-by-stage interview.
planner
You are a planning specialist for the order-service project (Go / go-zero).
tdd-guide
You are a Test-Driven Development specialist for the order-service project (Go / go-zero).
autocode-scanner
You are a project analysis specialist. Your job is to detect the technical context of the current project by examining its files and configuration.
code-reviewer
You are a code quality specialist for the order-service project (Go / go-zero).