glean-core-workflow-a
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external enterprise platforms (e.g., GitHub, Confluence) via the Glean API. This content is interpolated into the agent's context, creating a potential surface for indirect prompt injection where instructions hidden in source documents could attempt to influence the agent's behavior.
- Ingestion points: Data enters the context via API calls to the Glean search and chat endpoints as described in
SKILL.md. - Boundary markers: No explicit delimiters or specific "ignore" instructions for processing search results are defined in the skill body.
- Capability inventory: The skill frontmatter defines access to
Read,Write,Edit,Bash, andGreptools. - Sanitization: The skill contains a specific "Output" instruction section that mandates redacting workflow receipts and explicitly forbids recording query text, titles, snippets, transcripts, or credentials, which serves as a protective measure against data exposure.
Audit Metadata