recoup-internal-consulting-content-extraction
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external, untrusted content (meeting transcripts and notes) which could contain instructions designed to manipulate the agent's behavior.
- Ingestion points: Untrusted transcript data enters the agent context from files located in
content/01-raw/orclients/{client}/meetings/(SKILL.md). - Boundary markers: The instructions do not specify any delimiters or safety warnings to distinguish between control instructions and the data being processed.
- Capability inventory: The skill uses file-read and file-write capabilities to manage data within the workspace (SKILL.md).
- Sanitization: There is no evidence of sanitization or filtering applied to the transcript content before it is processed by the AI.
Audit Metadata