share-knowledge-to-memos
Pass
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [NO_CODE]: The skill consists entirely of markdown-based instructions and metadata. There are no scripts, binaries, or other executable files included.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process the current conversation history, which may contain untrusted content from external sources accessed during the session.
- Ingestion points: Step 1 ("Harvest") uses the conversation history as the source for all candidate conclusions.
- Boundary markers: No technical delimiters or specific instruction-data isolation prompts are defined.
- Capability inventory: The skill uses tools (
memo_list_memos,memo_create_memo,memo_update_memo) to search and write to an external database. - Sanitization: Step 5 ("Review") requires the user to manually review, edit, and approve all topic memos before any write operations occur, providing a strong human-in-the-loop defense against malicious data propagation.
Audit Metadata