apify-lead-generation

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

• Third-party content exposure detected (high risk: 0.75). Step 2 (“Fetch Actor Schema”) calls mcpc ... fetch-actor-details for a selected third-party Apify Actor, and the returned Actor description/README and parameter metadata are ingested as readable text into the agent context via jq -r ".content" (outsider-authored public web content from Apify at runtime).

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly runs mcpc against mcp.apify.com (mcpc --json mcp.apify.com ...) at runtime to fetch Actor details/README and input schemas that are then injected into the agent workflow to control prompts and runtime behavior, so mcp.apify.com is a required external source that can directly influence instructions.