audience-fit-check
Pass
Audited by Gen Agent Trust Hub on Jul 2, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill connects to the vendor's official API endpoint at
mcp.unifapi.comto fetch public profile data, posts, and engagement metrics from social media platforms. - [COMMAND_EXECUTION]: Documentation includes instructions for the user to install additional agent components using the
npxpackage runner from the vendor's repository. - [PROMPT_INJECTION]: The skill processes untrusted external data (social media posts and comments), which is a characteristic surface for indirect prompt injection.
- Ingestion points: Social media posts and comment threads fetched from social media platforms via the
unifapitool. - Boundary markers: Not explicitly defined in the workflow, though the skill requires citing specific posts for all findings to ensure data provenance.
- Capability inventory: Reading local project marketing files and calling external data APIs to fetch remote content.
- Sanitization: No content filtering or sanitization of social media data is mentioned before processing.
Audit Metadata