TheOneKit Security — STRIDE + OWASP Audit

Security audit combining STRIDE threat modeling with OWASP Top 10 checks. Produces actionable findings with severity levels and suggested fixes.

Usage

/t1k:security                          # Audit entire project
/t1k:security src/auth/                # Audit specific directory
/t1k:security --scope auth             # Focus on auth subsystem
/t1k:security --auto-fix               # Audit + offer to apply fixes
/t1k:security --report                 # Save findings to plans/reports/

STRIDE Categories

Category	Threat	Key Checks
Spoofing	Identity forgery	Auth bypass, session hijacking, token forgery
Tampering	Data modification	Input validation, SQL injection, XSS, CSRF
Repudiation	Denying actions	Audit logging gaps, missing attribution
Information Disclosure	Data leakage	Error messages, debug endpoints, PII in logs
Denial of Service	Availability attack	Rate limiting, ReDoS, large payload
Elevation of Privilege	Access escalation	RBAC bypass, mass assignment, insecure defaults

Full per-category checks and OWASP Top 10 mapping: references/stride-checks.md

Audit Process

1. Determine scope (full project or --scope flag)
2. For each STRIDE category: run checks from references/stride-checks.md
3. Assign severity (Critical/High/Medium/Low/Info)
4. Group findings by category
5. Output report in standard format (see references/stride-checks.md)

Flags

• --auto-fix: After reporting, list High+Critical with code changes. Confirm per-fix before applying — never bulk-apply.
• --report: Save findings to plans/reports/security-audit-{date}.md
• --scope auth|api|data: Limit audit to subsystem

Gotchas

• False positives on crypto: Library wrappers may look like raw crypto — check the wrapper implementation before flagging
• Config-based auth: Some frameworks apply auth globally via middleware config — check config files, not just route handlers
• SSRF via URL params: Any endpoint accepting a URL parameter is an SSRF candidate — always flag for review
• Logging PII: Check both explicit log statements AND error serialization (stack traces may include request bodies)
• ReDoS detection: Look for regex patterns with nested quantifiers: (a+)+, (a|a)*, ([a-z]+)* — dangerous

Auto-Activation Keywords

Triggers on: security audit, security review, STRIDE, OWASP, threat model, vulnerability, injection, XSS, CSRF, authentication review, authorization check, pentest, security scan