auditor
Skill2 starsChanged todayMITView on GitHub
Reviews code for security vulnerabilities, dependency risks, and access control issues. Use before merging any security-sensitive change, on a regular audit schedule, or when adding new dependencies. The Auditor assumes breach and reads code the way an attacker would.
- Reads credentials
6 other projects use a version of this.
Adopted by
Most recently taken up by
- NotRedFox/NotRedFoxs-Claude-skillstheir own version, first seen Oct 9, 2026
- opentensor/subtensortheir own version, first seen Oct 8, 2026
- edisonzerolam/openclaw-skillstheir own version, first seen Jul 4, 2026
- QinghongLin/data2story-skilltheir own version, first seen Jul 2, 2026
- daemon-blockint-tech/Agentic-Enteprises-Skilltheir own version, first seen May 20, 2026
- rade-nugroho/Agentic-Enteprises-Skilltheir own version, first seen May 20, 2026
Other versions6 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 6
- opentensor/subtensor★ 402same name
- QinghongLin/data2story-skill★ 155same name
- rade-nugroho/Agentic-Enteprises-Skill★ 9same name
- daemon-blockint-tech/Agentic-Enteprises-Skill★ 8same name
- NotRedFox/NotRedFoxs-Claude-skills★ 1same name
- edisonzerolam/openclaw-skills★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository60 other files it gives its agents.
- AGENTS.mdAGENTS.md
- accessibility-auditorSkill
- architectSkill
- awsSkill
- bug-fix-teammateSkill
- builderSkill
- cleanup-specialistSkill
- code-reviewSkill
- commit-messagesSkill
- concept-explainerSkill
- context-economySkill
- cross-provider-translationSkill
- datadogSkill
- debuggerSkill
- debugging-and-error-recoverySkill
- debugging-tutorSkill
- dockerSkill
- doormanSkill
- elasticsearchSkill
- emailSkill
- envoySkill
- githubSkill
- gitlabSkill
- goal-refinementSkill
- heraldSkill
- implementation-plannerSkill
- inspectorSkill
- institutional-knowledgeSkill
- issue-managerSkill
- jiraSkill
- kubernetesSkill
- librarianSkill
- linearSkill
- mailmanSkill
- mediatorSkill
- mongodbSkill
- mysqlSkill
- onboarding-planSkill
- operatorSkill
- oracleSkill
- postgresSkill
- pull-request-assistantSkill
- redisSkill
- release-notesSkill
- rememberSkill
- reviewerSkill
- runtime-healthSkill
- scribeSkill
- security-and-hardeningSkill
- sentinelSkill
- sentrySkill
- society-integritySkill
- spec-driven-developmentSkill
- stewardSkill
- strategistSkill
- telegramSkill
- test-driven-developmentSkill
- testerSkill
- validation-and-enforcementSkill
- wardenSkill
The file
We split this skill into the rules it gives an agent
10 rules. 0 are common, 7 are distinctive, and 3 are very specific to this project.
0
Common
Most of the 5 edited versions we read keep these.
7
Distinctive
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 16
Before merging any change that touches auth, user input, or data persistence
- Distinctive: line 17
When adding new dependencies
- Distinctive: line 19
When a security advisory is published for a used dependency
- Distinctive: line 20
When a new API endpoint or data access pattern is introduced
- Very specific: line 58
Run npm audit or equivalent — are there known CVEs in dependencies?
- Distinctive: line 80
If yes: is it validated against an allowlist of permitted hosts?
- Distinctive: line 97
Before any commit is finalized, scan staged changes for:
- Distinctive: line 104
If found: block the commit, instruct to remove from history, rotate the exposed credential immediately.
- Very specific: line 153
Use GITLEAKS_LICENSE: ${{ secrets.GITLEAKS_LICENSE || '' }} — works on personal accounts today, ready for org transfer without workflow …
- Very specific: line 155
Repo visibility detection: Use github.event.repository.private in workflow conditions to warn (not fail) when the repo is private and no …
“Common” is measured: kept by at least 6 in 10 of the 5 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
The Auditor
Overview
The Auditor assumes breach. It reads code the way an attacker would. It does not care that the input "will never be null" or that the endpoint "is only called internally." It verifies. It does not trust that the dependency "is probably fine." It checks. It is not paranoid — it is precise.
When to Use
- Before merging any change that touches auth, user input, or data persistence
- When adding new dependencies
- On a scheduled audit cadence (weekly or per release)
- When a security advisory is published for a used dependency
- When a new API endpoint or data access pattern is introduced
Process
OWASP Top 10 Systematic Review
Work through each risk category for every changed file:
A01 — Broken Access Control
- Is every protected route/endpoint checking authentication?
- Is every protected resource checking authorization (not just authentication)?
- Are access control checks server-side, not just client-side?
- Are direct object references (IDs) validated against the current user's permissions?
A02 — Cryptographic Failures
- Are secrets stored in environment variables, not source code?
- Are passwords hashed with a strong algorithm (bcrypt, argon2) — not MD5 or SHA1?
- Is sensitive data encrypted at rest and in transit?
- Are TLS certificates valid and enforced?
A03 — Injection
- Are all SQL queries parameterized? (Zero string concatenation with user input)
- Is user input used in shell commands? (Must never be)
- Is user input used in file paths? (Must be sanitized and validated)
- Are template engines escaping output by default?
A04 — Insecure Design
- Is there a trust boundary between authenticated and unauthenticated zones?
- Are rate limits in place on authentication endpoints?
- Is sensitive functionality (delete, admin actions) behind additional confirmation?
A05 — Security Misconfiguration
- Are CORS origins explicit (not
*) in production? - Are error messages revealing stack traces or internal details to users?
- Are default credentials changed?
- Are unnecessary features and endpoints disabled?
A06 — Vulnerable Components
- Run
npm auditor equivalent — are there known CVEs in dependencies? - Are dependencies using wildcard versions (
*,^latest)? - Are any dependencies abandoned (no release in 2+ years)?
A07 — Authentication Failures
- Are session tokens sufficiently random and long?
- Are failed login attempts rate-limited?
- Is session invalidation happening on logout?
- Are password reset tokens single-use and time-limited?
A08 — Software and Data Integrity Failures
- Are dependencies installed from trusted registries with lockfiles committed?
- Is deserialization of untrusted data avoided?
- Are CI/CD pipeline configurations protected from unauthorized modification?
A09 — Logging and Monitoring Failures
- Are authentication events (login, logout, failure) logged?
- Are the logs free of sensitive data (passwords, tokens, PII)?
- Are logs immutable and retained for an appropriate duration?
A10 — Server-Side Request Forgery (SSRF)
- Is any user-supplied URL used to make a server-side HTTP request?
- If yes: is it validated against an allowlist of permitted hosts?
Dependency Audit
For every new dependency added:
- Necessity check — does the existing stack already solve this?
- Size check — what is the bundle/install size impact?
- Maintenance check — last release date, open issues, contributor activity
- Vulnerability check —
npm auditorpip auditfor known CVEs - License check — is the license compatible with the project?
- Transitive check — what does this dependency bring in?
Flag any dependency that fails two or more checks.
Secret Scanning
Before any commit is finalized, scan staged changes for:
- API keys (patterns:
sk_,pk_,key_,secret,token,password) - Connection strings with embedded credentials
.envfiles accidentally staged- Private keys and certificates (
-----BEGIN) - AWS credentials (
AKIA,aws_access_key)
If found: block the commit, instruct to remove from history, rotate the exposed credential immediately.
Blocking Findings
The following are always [blocking] — they prevent merge regardless of urgency:
- Hardcoded secrets or API keys in any committed file
- SQL queries built with string concatenation of user input
dangerouslySetInnerHTMLwithout explicit sanitization- Authentication checks missing on protected endpoints
- Dependencies with critical or high CVEs without a mitigation plan
- User input used directly in shell command execution
Red Flags
- "This endpoint is internal only" — internal endpoints are still attack surfaces
- "We'll add auth later" — auth is not a feature, it is a foundation
- New dependency with no lockfile update
- Error responses that include stack traces
- Logging that captures request bodies (may contain passwords)
- CORS set to
*anywhere except a public static file server
Rationalizations
| What you think | What The Auditor knows |
| "This will never be called with malicious input" | Every endpoint that exists can be called with malicious input. |
| "We're not big enough to be targeted" | Automated scanners do not care about your size. |
| "The dependency is popular so it must be safe" | Popular dependencies are popular targets. Popularity is not a security audit. |
| "We'll do a security review before launch" | Security is not a phase. It is built in, not bolted on. |
Verification
Audit is complete when:
- All OWASP Top 10 categories checked for changed files
- No hardcoded secrets in staged or committed changes
- All new dependencies audited for CVEs, license, and maintenance
- All
[blocking]findings are resolved - Auth checks verified on all new or modified endpoints
- Logging reviewed for sensitive data leakage
Implementation Notes (CI Secret Scanning)
Canonical action: gitleaks/gitleaks-action@v2 (migrated from zricethezav/gitleaks-action)
License requirements:
- Personal GitHub accounts: no license needed —
GITLEAKS_LICENSEcan be omitted or empty - Organization accounts: free Starter license required (one repo) from gitleaks.io
- Use
GITLEAKS_LICENSE: ${{ secrets.GITLEAKS_LICENSE || '' }}— works on personal accounts today, ready for org transfer without workflow changes
Repo visibility detection: Use github.event.repository.private in workflow conditions to warn (not fail) when the repo is private and no license secret is set, rather than silently producing incorrect results.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

