Awesome-omni-skill security-audit
Use when reviewing code for security vulnerabilities and OWASP top 10 issues
install
source · Clone the upstream repo
git clone https://github.com/diegosouzapw/awesome-omni-skill
Claude Code · Install into ~/.claude/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/diegosouzapw/awesome-omni-skill "$T" && mkdir -p ~/.claude/skills && cp -r "$T/skills/testing-security/security-audit-renatinhosfaria" ~/.claude/skills/diegosouzapw-awesome-omni-skill-security-audit-c77359 && rm -rf "$T"
manifest:
skills/testing-security/security-audit-renatinhosfaria/SKILL.mdsource content
Security Audit - Vulnerability Assessment
The Iron Law
ASSUME ALL INPUT IS MALICIOUS. Every system boundary is an attack surface. Every data flow from an untrusted source must be validated, sanitized, and escaped before use.
Process
- Identify Attack Surface -- Map all entry points (HTTP params, file uploads, DB reads, env vars, third-party APIs, WebSocket messages)
- Injection Audit -- Check for SQL, command, XSS, NoSQL, and template injection (see references for full table)
- Auth & Session Audit -- Verify password hashing, session invalidation, JWT validation, rate limiting
- Authorization Audit -- Confirm AuthZ on every endpoint, tenant isolation, server-side role checks
- Data Exposure Audit -- Check for hardcoded secrets, sensitive data in logs, missing encryption
- Dependency Audit -- Run
and check for outdated packages with known CVEspnpm audit - Categorize Findings -- Rate each finding by severity (Critical/High/Medium/Low/Info)
Checklist
- Attack surface mapped (all entry points identified)
- Injection audit completed (SQL, command, XSS, NoSQL, template)
- Authentication and session audit completed
- Authorization audit completed (every endpoint checked)
- Sensitive data exposure audit completed
- Dependency audit run (pnpm audit)
- Findings categorized by severity
- Report generated in structured format
Red Flags
STOP the audit and escalate immediately if you find:
- Hardcoded credentials (API keys, passwords, database URIs) in source code
- Disabled authentication or authorization checks
- SQL queries built with string concatenation from user input
- Sensitive data (passwords, tokens, PII) written to logs
- CORS set to
in production configuration* - Default or weak credentials in production environment