meeting-distiller-pro
Pass
Audited by Gen Agent Trust Hub on Oct 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input from meeting transcripts and chat logs.
- Ingestion points: Untrusted data enters the context via transcripts, notes, and chat logs (SKILL.md).
- Boundary markers: No structural delimiters (like XML tags) are specified for the input data, though the instructions include a warning to treat transcript content as untrusted.
- Capability inventory: The skill has no capabilities; it does not use tools, perform network operations, access the file system, or execute commands.
- Sanitization: No sanitization of the input text is performed, but the lack of capabilities prevents exploitation.
Audit Metadata