gws-keep
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from Google Keep notes via the
notesandmediaresources. If a note contains malicious instructions, they could influence the agent's behavior during the 'get' or 'list' operations. Specifically, thenotes getandnotes listmethods retrieve note content that may contain untrusted data, although the risk is mitigated by standard LLM guardrails.
Audit Metadata