compound
Pass
Audited by Gen Agent Trust Hub on Nov 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill extracts information from untrusted task outcomes and project deliverables to build a persistent knowledge base.
- Ingestion points: Processes task results, milestone outcomes, and unexpected events as described in Step 1 of the workflow.
- Boundary markers: The instructions do not define delimiters or provide 'ignore embedded instructions' warnings for the data being processed.
- Capability inventory: The skill utilizes /context-save and manual file updates to persist extracted learnings into context/learnings/ and context/index.json.
- Sanitization: No validation or sanitization of the extracted learnings is specified before they are committed to the filesystem.
Audit Metadata