job-posting-signals
Pass
Audited by Gen Agent Trust Hub on Sep 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill directs the agent to ingest and analyze data from external websites like company careers pages and job boards. \n- Ingestion points: Career pages, job boards, and LinkedIn Jobs. \n- Boundary markers: Absent; no instructions to ignore embedded instructions in the scraped data. \n- Capability inventory: Live web searching and data extraction of role details and pain sentences. \n- Sanitization: Absent; no filtering or validation of external content. \n- [EXTERNAL_DOWNLOADS]: The skill provides a setup command to install an external Model Context Protocol (MCP) server from the vendor. \n- Evidence:
claude mcp add --transport http manypi https://mcp.manypi.com/mcp -s userin SKILL.md. \n- Description: This configuration adds the vendor's tool to the agent's environment to enable real-time search capabilities.
Audit Metadata