teach

Pass

Audited by Gen Agent Trust Hub on Jul 23, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill recommends using CLI commands to open generated HTML lesson files for the user. This is a localized helper action intended for workflow convenience and does not involve executing untrusted remote code.\n- [PROMPT_INJECTION]: The skill incorporates a surface for indirect prompt injection as it reads from various local files in the workspace. Ingestion points: MISSION.md, RESOURCES.md, NOTES.md, and files in the learning-records directory. Boundary markers: Absent. Capability inventory: Writing to the file system and executing commands to open local files. Sanitization: Not specified. While this creates a potential surface for processing untrusted data, the skill's own instructions are non-malicious and focus on pedagogical goals.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 23, 2026, 08:19 AM
Security Audit — agent-trust-hub — teach