memory-start
Pass
Audited by Gen Agent Trust Hub on Nov 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill implements a persistent memory system that stores user-provided content via the
bm_writetool and retrieves it in future sessions usingbm_searchandbm_edit, creating a surface for indirect prompt injection. - Ingestion points: User-provided descriptions of their work and status are captured in the
contentparameter of thebm_writetool. - Boundary markers: The instructions do not specify the use of delimiters or protective headers to prevent the agent from following instructions potentially embedded within the stored notes.
- Capability inventory: The skill has the ability to list projects, search notes, write new notes, edit existing notes, delete notes, and open notes in an external viewer.
- Sanitization: There is no evidence of content sanitization, validation, or filtering of the user data before it is persisted or subsequently processed.
Audit Metadata