signal-sourcer

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

• Third-party content exposure detected (high risk: 0.75). The skill’s runtime workflow explicitly calls the ColdIQ MCP search functions (mcp__coldiq__search_web / mcp__coldiq__find_signals), which can fetch and ingest public web content (outsider-authored) into the agent’s LLM context via the tool output.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill's setup instructs the operator to run a runtime fetch-and-execute command ("COLDIQ_API_KEY= npx -y @coldiq/mcp@latest") tied to ColdIQ (see https://coldiq.com/marketplace), which downloads and runs remote code during the session.