Skip to content

MustafaKucukcoskun/Refine-agent-kit / .agents/skills/trail-of-bits-security

trail-of-bits-security

Verified

Trail of Bits Security Principles

This skill is global and should be applied in every domain.

1. Input Validation

  • Treat all data coming from outside the system (User, API Request, File) as "MALICIOUS" by default.
  • Prefer Validation over Sanitization (e.g., is_integer, Pydantic strict types).
  • Always perform path normalization (e.g., os.path.abspath) and prefix checks for file paths (../../) to prevent path traversal vulnerabilities.

2. Secrets Management

  • Never hardcode secrets, tokens, or API/DB connection strings, even for testing purposes.
  • Use .env, and always assume that prefixes considered safe like EXPO_PUBLIC_ leak to the browser/mobile (do not put real secrets in them).
  • Ensure critical keys remain outside source control (Git) (e.g., .gitignore).

3. OWASP Top 10 and Injection

  • Always use Parameterized Queries or ORM for SQL data retrieval. String concatenation ("SELECT * FROM users WHERE name = " + user_input) is strictly forbidden.
  • Do not render user input without HTML encoding for XSS protection.
  • Ensure Security Headers like X-Frame-Options, CSP, HSTS are served from the API or Frontend server.

4. AuthZ (Authorization) and AuthN (Authentication)

  • Authentication only identifies who the user is; it does not determine their permissions.
  • Authorization must be independently checked at the beginning of every endpoint. (Checking permissions only at the root route is insufficient; pay attention to object/row-level permissions - Insecure Direct Object Reference (IDOR) risk).
  • If using JWT or OAuth, do not create very long-lived Access Tokens without a Refresh Token.

5. Dependency and Log Security

  • Regularly update third-party dependencies (consider tools like npm audit, pip-audit, snyk, or trivy in your project).
  • When catching exceptions and writing to logs, never output user PII (Personal Identifiable Information), passwords, national IDs, Session IDs, or bearer token values in plaintext.
rea-analysis-optimizationmorluto/reaOptimize REA providers and shared analysis workflows for agent context, latency, CPU, and memory while preserving correctness and Evidence. Use for analysis bottlenecks and cache or query design; CI/test-suite optimization belongs to rea-ci-tests-audit.Stars: 471kUpdated Claude CodeCodexVerifiedInstall
rea-ci-tests-auditmorluto/reaAudit REA CI and test suites for correctness, fixture drift, false passes, and measured performance; use for suite or workflow investigations and efficiency changes, rather than ordinary feature testing or a single test failure.Stars: 471kUpdated Claude CodeCodexVerifiedInstall
rea-tool-designmorluto/reaAudit, investigate, design, or change REA investigation tools, CLI/MCP contracts, provider capabilities, and Evidence semantics.Stars: 471kUpdated Claude CodeCodexVerifiedInstall
ponytail-auditDietrichGebert/ponytailQuality audit of the whole repo: bugs, security, real load, missing tests, speed, and what to delete. Most important first.Stars: 194kUpdated Claude CodeCodexVerifiedInstall