Pass
Audited by Gen Agent Trust Hub on Jul 3, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is a 'Data Skill' designed strictly for read-only access to public LinkedIn data through UnifAPI's MCP server. It does not contain any executable code or scripts, reducing the risk of runtime attacks.
- [SAFE]: No remote code execution or external downloads were detected. The skill references legitimate vendor domains (unifapi.com, github.com/unifapi-agent) which are consistent with the skill's authorship.
- [SAFE]: Data exposure and exfiltration risks are mitigated by the architecture; the skill uses OAuth for authentication managed by the UnifAPI platform, and instructions explicitly state it never holds credentials or sends messages.
- [SAFE]: No obfuscation (Base64, homoglyphs, or zero-width characters) was found in either the SKILL.md or README.md files.
- [SAFE]: The skill uses clear instructions for agent behavior, focusing on deterministic data retrieval recipes rather than trying to override safety filters or system prompts.
Audit Metadata