best-practices-python

Pass

Audited by Gen Agent Trust Hub on Mar 17, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill serves as a documentation and linting suite for Python development. It explicitly promotes secure coding practices by including rules against command injection (security-no-shell-true.md), code injection (security-no-eval-exec.md), and credential leakage (security-redact-secrets.md).
  • [COMMAND_EXECUTION]: The skill includes shell and Python scripts for project maintenance and verification. Scripts like 'sanity.sh' and 'sanity/lint_dotenv.sh' use standard tools (find, grep, python3) to ensure project compliance. 'scripts/check_file_limits.py' and 'analyze_packages.py' scan the local file system for Python files and dependency configurations. These operations are restricted to the local workspace and align with the skill's stated purpose.
  • [DATA_EXFILTRATION]: No network communication or data transmission patterns were detected. Scripts only perform local file I/O for report generation and linting purposes.
  • [PROMPT_INJECTION]: Analysis of the markdown content and metadata found no instructions attempting to override agent behavior, bypass safety protocols, or extract system prompts.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 17, 2026, 06:34 AM
Security Audit — agent-trust-hub — best-practices-python