meeting-insights-analyzer
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted meeting transcripts, which could contain instructions to influence the agent's behavior.\n- Ingestion points: Meeting transcripts in formats such as .txt, .md, .vtt, .srt, and .docx are read from local directories as instructed in SKILL.md.\n- Boundary markers: The instructions in SKILL.md do not define specific delimiters or instructions to ignore embedded commands within the transcripts.\n- Capability inventory: Analysis of SKILL.md reveals no subprocess calls, dynamic execution (exec/eval), file-writing capabilities, or network operations requested by the skill instructions.\n- Sanitization: No logic for sanitizing, escaping, or validating the content of the meeting transcripts is described in SKILL.md.\n- [SAFE]: The skill operates locally on user-provided data and follows standard instructional patterns for a text analysis tool without introducing security risks.
Audit Metadata