source-management
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to search and process data from various external sources including chat, email, cloud storage, CRM, and project trackers. This creates an attack surface for indirect prompt injection where malicious instructions embedded in these external data sources could influence the agent's behavior during search and digest tasks.
- Ingestion points: External search results from tools prefixed with
~~chat,~~email,~~cloud storage,~~project tracker,~~CRM, and~~knowledge baseinSKILL.md. - Capability inventory: The skill facilitates searching, reading messages/files, and aggregating data across enterprise platforms.
- Boundary markers: The skill lacks instructions for using delimiters or boundary markers to separate retrieved untrusted data from the agent's instruction context.
- Sanitization: No explicit sanitization, filtering, or escaping logic is defined for the content ingested from external platforms before it is processed by the agent.
Audit Metadata