workorai

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.30). The WorkorAI talent-marketplace skill allows employers to search for and review candidates via MCP tools like employer.search_candidates_for_job and employer.get_applicant_transcript, which return candidate-authored profiles, resumes, and interview responses containing untrusted outsider free text.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.70). The skill instructs connection to and execution of a remote Model Context Protocol (MCP) server at https://workorai.com/mcp from a community-sourced repository (work0r-ai/agent-kit), constituting an unverified third-party runtime dependency.