memory-research
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from external web searches and interpolates it into prompts for summarization and note creation.
- Ingestion points: External web content retrieved via search tools in Step 1.
- Boundary markers: The instructions lack explicit delimiters or warnings to ignore malicious instructions embedded in the retrieved web data.
- Capability inventory: Uses write_note and edit_note to modify a local knowledge base.
- Sanitization: No explicit sanitization or filtering of external content is mentioned.
- Mitigation: The workflow includes a mandatory step to present findings and seek user approval before creating or editing notes.
Audit Metadata