claude-code-best-practice

Pass

Audited by Gen Agent Trust Hub on May 16, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is entirely instructional, providing guidance and examples for structuring Claude Code agents, commands, and workflows. No hidden content or malicious patterns were detected.
  • [COMMAND_EXECUTION]: Provides example shell commands for standard tasks such as package installation (pip install), permission management (chmod +x), and environment configuration. All examples use standard developer tools and placeholders for sensitive variables.
  • [EXTERNAL_DOWNLOADS]: References official and well-known repositories, including the Model Context Protocol (MCP) servers on NPM and documentation repositories on GitHub. These are legitimate resources for the skill's stated purpose.
  • [CREDENTIALS_UNSAFE]: Correctly uses environment variable placeholders (e.g., ${GITHUB_TOKEN}, ${ANTHROPIC_API_KEY}) for sensitive configuration, following security best practices for credential management.
Audit Metadata
Risk Level
SAFE
Analyzed
May 16, 2026, 02:30 PM
Security Audit — agent-trust-hub — claude-code-best-practice