install-legal-data-hunter

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

• Third-party content exposure detected (high risk: 0.85). The skill’s runtime workflow connects to an external MCP server (via claude mcp add ... https://legaldatahunter.com/mcp or npx legal-data-hunter --yes), and the agent can call tools like get_document(id) / search(...) that return full text authored by third parties (case law/legislation/doctrine), which is then ingested into the agent’s LLM context.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill instructs at runtime to connect to and load tools from the external MCP endpoint (https://legaldatahunter.com/mcp) via claude mcp add ... and to execute remote code via npx legal-data-hunter --yes, both of which fetch and run external content that can directly control agent prompts and behavior.