docs-explorer

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
  • [Indirect Prompt Injection] (LOW): The skill retrieves content from external sources (brain repos and Notion), which provides a surface for indirect prompt injection.\n
  • Ingestion points: Brain repository paths and Notion page content specified in SKILL.md.\n
  • Boundary markers: Absent; there are no instructions to the agent to distinguish between its own logic and instructions found within documents.\n
  • Capability inventory: No code execution or file-write permissions are defined in this skill.\n
  • Sanitization: Absent; content is synthesized into the output format without validation.\n- [No Code] (SAFE): The skill definition contains only natural language and markdown templates, with no executable logic or scripts.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:42 PM