find-people
Pass
Audited by Gen Agent Trust Hub on Nov 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external content from sources like LinkedIn profiles, engineering blogs, and GitHub contributor bios, which could contain malicious instructions designed to influence the agent's behavior.
- Ingestion points: Web search results and LinkedIn profile content fetched via browser tools (referenced in SKILL.md and resources/linkedin-browser.md).
- Capability inventory: The skill aggregates professional information and hands off data to the
draft-outreachskill; it does not explicitly use subprocess execution, file writing, or network exfiltration tools. - Boundary markers: Instructions specify using source citations and marking uncertain information as unverified, which provides a degree of transparency.
- Sanitization: There is no evidence of explicit sanitization or filtering to prevent the agent from obeying instructions embedded in the external content it reads.
Audit Metadata