reply-io-automation

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

• Third-party content exposure detected (high risk: 0.70). This skill requires calling RUBE_SEARCH_TOOLS against the public MCP endpoint (e.g., https://rube.app/mcp and composio.dev/toolkits/reply_io) to fetch live tool schemas, recommended execution plans, and tool slugs from third-party providers, which the agent must read and act on to construct and execute tool calls, allowing untrusted external content to influence behavior.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill requires contacting the Rube MCP server at https://rube.app/mcp at runtime to fetch tool schemas and recommended execution plans that directly determine which tools/instructions the agent will execute, so this external endpoint can control the agent's prompts and actions.