who-is-this
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external platforms (GitHub, LinkedIn, and X) to assess a person's credibility.
- Ingestion points: Data is ingested from user-provided handles, URLs, and profile screenshots, which are then processed via the deepapi tool for scraping profiles and posts.
- Boundary markers: There are no explicit delimiters or instructions for the agent to ignore control-related content within the scraped text or images.
- Capability inventory: The skill leverages the deepapi capability to perform automated research and scraping.
- Sanitization: The instructions lack technical sanitization, relying instead on high-level goals. An adversary could include hidden prompts in their public-facing profiles to steer the agent's archetype classification or credibility assessment.
Audit Metadata