countdown-api-automation

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

• Third-party content exposure detected (high risk: 0.80). The skill instructs the agent to call RUBE_SEARCH_TOOLS against Rube MCP (e.g., https://rube.app/mcp / composio.dev/toolkits/countdown_api) to fetch external tool slugs, schemas, recommended execution plans and pitfalls, which the agent must read and act on—exposing it to untrusted third-party content that could carry indirect instructions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill instructs the agent to use https://rube.app/mcp as the Rube MCP server and relies on RUBE_SEARCH_TOOLS / RUBE_MULTI_EXECUTE_TOOL runtime calls to that external endpoint to discover and execute tools, which means that fetched content from that URL can directly control prompts/instructions or execute remote code at runtime.