meeting-insights-analyzer
Pass
Audited by Gen Agent Trust Hub on Mar 13, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes untrusted meeting transcripts which creates a surface for indirect prompt injection attacks. 1. Ingestion points: transcript files (.txt, .md, .vtt, .srt, .docx) referenced in SKILL.md. 2. Boundary markers: No delimiters are specified to isolate transcript content from instructions. 3. Capability inventory: Limited to text analysis, statistics calculation, and summary generation as described in SKILL.md. 4. Sanitization: No sanitization or validation of the transcript content is mentioned.
- [SAFE]: No evidence of hardcoded credentials, sensitive file access, or unauthorized data transmission was found.
- [SAFE]: The skill does not download or execute remote code and has no dependencies on external scripts.
- [SAFE]: No obfuscation or hidden persistence mechanisms were detected in the skill instructions.
Audit Metadata