meeting-insights-analyzer
Pass
Audited by Gen Agent Trust Hub on Oct 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external meeting transcript files (.txt, .md, .vtt, .srt, .docx), which constitutes an untrusted data source. A malicious actor could theoretically embed instructions within a transcript to manipulate the agent's behavioral analysis or summary.
- Ingestion points: Instructions specify scanning folders for multiple transcript formats (.txt, .md, .vtt, .srt, .docx) and reading their content.
- Boundary markers: The instructions lack explicit delimiters or "ignore embedded instructions" directives for the data processing phase.
- Capability inventory: The skill's functionality is limited to reading files and generating text-based analysis; it does not utilize network operations, file-writing tools, or system command execution.
- Sanitization: There are no documented steps for sanitizing or filtering the content retrieved from the transcripts before analysis.
Audit Metadata