agentsclimarketplace

Security audit

Skill onfire7777/universal-ai-skills-library/skills/security-audit

Router-first AI skill system for Codex, Claude, Cursor, Hermes, Paperclip, OpenCode, and local AI stacks: search, preflight-route, and load 1,812 skills on demand without duplicating the corpus.

Install
npx -y skills add onfire7777/universal-ai-skills-library --skill security-audit

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 13 stars13 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Offensive security audit — think like an attacker, report like a defender. OWASP Top 10 2025, secrets detection, dependency CVEs, injection testing. Use when reviewing code security, preparing for penetration tests, or auditing a project before launch. Triggers on: 'security audit', 'vulnerability scan', 'is this secure', 'check for vulnerabilities', 'OWASP', 'penetration test', 'security review', 'find security issues'.

SKILL.md

2.1 KB, 395 tokens by cl100k_base, as published. Nobody here has run it

Security Audit

Thinking Protocol

Before auditing, answer silently:

  1. What is the attack surface? (public endpoints, user inputs, file uploads, auth flows)
  2. What data is most valuable to an attacker?
  3. What is the blast radius if compromised?

Execution

Phase 1: Reconnaissance

  • Map attack surface: entry points, data flows, trust boundaries
  • Identify stack and known CVEs per dependency
  • Scan for secrets: API keys, tokens, passwords, connection strings

Phase 2: OWASP Top 10 (2025) Sweep

Actively attempt to find vulnerabilities for each category:

  1. Broken Access Control — privilege escalation paths
  2. Cryptographic Failures — plaintext storage, weak algorithms
  3. Injection — SQL, XSS, SSRF, command, path traversal
  4. Insecure Design — missing rate limits, threat modeling gaps
  5. Security Misconfiguration — debug mode, open CORS, verbose errors
  6. Vulnerable Components — known CVEs in dependencies
  7. Auth Failures — weak passwords, session fixation, JWT issues
  8. Data Integrity — unsigned updates, CI/CD pipeline poisoning
  9. Logging Gaps — missing audit trails, PII in logs
  10. SSRF — internal service access from server

Phase 3: Report

Per finding:

[CRITICAL|HIGH|MEDIUM|LOW] — [Vulnerability Type]
Location: [file:line or endpoint]
Attack: [How exploited — specific]
Impact: [What attacker gains]
Fix: [Exact code/config change]

Priority-ordered remediation plan at end.

Rules

🚨 Think like an attacker, not an auditor. Try to break things. 🚨 Lead with critical findings. Don't bury them under low-severity noise. 🚨 No false positives. Verify before reporting.

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.