meeting-notes

Pass

Audited by Gen Agent Trust Hub on Sep 21, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data in the form of meeting transcripts and notes.\n
  • Ingestion points: Processes transcripts or raw notes provided in the prompt context (SKILL.md).\n
  • Boundary markers: No specific delimiters or boundary markers are defined to isolate input text from instructions.\n
  • Capability inventory: The skill generates text and markdown output and does not contain subprocess calls, network operations, or file-writing capabilities.\n
  • Sanitization: The skill provides a mandatory "AI-transcript verification checklist" and explicit instructions to flag suspect content rather than asserting it as fact (SKILL.md, references/templates.md).
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 21, 2026, 11:11 PM
Security Audit — agent-trust-hub — meeting-notes