soia-pkm-bootstrap

Pass

Audited by Gen Agent Trust Hub on Jul 14, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill uses a local Python script (scripts/init_vault.py) to handle file system operations. The script is configuration-driven, reading from scripts/default_config.json to create directories and write static content. No dangerous dynamic execution or arbitrary shell commands were found.
  • [EXTERNAL_DOWNLOADS]: The documentation suggests installing related skills using npx skills add soia-team/soia-open-skills. These resources originate from the same vendor ('soia-team') and are part of the intended functionality for the PKM ecosystem.
  • [COMMAND_EXECUTION]: The bootstrapping process involves running a Python script to initialize the vault. The script's logic is restricted to creating folders and files at a user-specified path.
  • [SAFE]: The skill includes instructions for optional session logging (Step 6) but explicitly states that this must not be done silently and requires user consent, demonstrating a good security posture.
  • [SAFE]: The markdown templates generated for AI adapters (e.g., CLAUDE.md, GEMINI.md) include explicit safety instructions, such as 'Do not read private data or credential files unless the user explicitly asks.'
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 14, 2026, 03:44 PM
Security Audit — agent-trust-hub — soia-pkm-bootstrap