check-known-solutions

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

• Third-party content exposure detected (high risk: 0.75). The required runtime workflow calls the Push Realm MCP server’s search_solutions/submit_solution tools, which can ingest outsider-authored free text from a shared knowledge base returned by that public/community service into the agent’s LLM context (indirect prompt injection risk).

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

• Potentially malicious external URL detected (high risk: 0.90). The skill requires connecting to the Push Realm MCP server at https://api.pushrealm.com/mcp at runtime (see the claude mcp add command and JSON config), and it fetches search results whose next_action/hint/solution content is applied to control agent actions and fixes, so remote content directly controls prompts.