foundation-meeting-recap
Pass
Audited by Gen Agent Trust Hub on Oct 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external sources (Zoom, Otter, Meet transcripts) and writes extracted decisions to a local state file (
.claude/pm-skills.local.md). It lacks explicit boundary markers or sanitization to prevent the agent from following instructions embedded within the transcripts. - Ingestion points: Processes transcripts and notes from multiple sources as primary inputs (defined in SKILL.md under "Instructions").
- Boundary markers: No delimiters or "ignore embedded instructions" warnings are specified for the input parsing stage.
- Capability inventory: The skill writes decisions and artifacts to the local file system via the "Project Memory Contract" (.claude/pm-skills.local.md).
- Sanitization: Lacks explicit sanitization or filtering logic for the input text beyond a general fabrication prohibition.
Audit Metadata