global-labor-employment-data

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

• Third-party content exposure detected (high risk: 0.75). The required runtime workflow for query_labor_data pulls labor statistics from the World Bank WDI dataset (vendor/public data) and returns it as JSON for the agent to read, which is outsider-authored content entering the LLM context as readable indicator text.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill explicitly instructs the agent to call the AgentPMT MCP server (https://api.agentpmt.com/mcp/) using tools/call (e.g., get_schema/get_instructions) at runtime to fetch live schema and instructions, and that fetched content would directly control prompts and agent behavior.