{"name":"Runtype MCP Server","version":"0.1.0","protocol":"mcp","protocolVersion":"2025-11-25","description":"Model Context Protocol server for Runtype API","transport":"streamable-http","capabilities":{"tools":{},"resources":{},"prompts":{},"events":{},"extensions":{"io.modelcontextprotocol/events":{"listChanged":false}}},"instructions":"When the user wants to build an AI product, design or run a flow, configure a surface, embed a Persona chat widget, or manage agents, records, logs, or runtime tools on Runtype, use this server. Before building, call `get_build_instructions`: `build-product`, `generate-flow`, or `explain-capabilities` for scoping. Read-only account operations do not need build instructions. Use `get_platform_documentation` for schema, type-definition, tool-catalog, embed, SDK, or dashboard-link lookups, and `search_documentation` for prose/how-to questions answered from the docs corpus. For agents running outside Runtype (Flue, Cloudflare Agents SDK, Vercel AI SDK, LangGraph, OpenAI Agents SDK, Mastra, OpenClaw, Hermes, Pi, custom loops), read the `external-agents` documentation topic for native telemetry plugins, callable endpoints, and eval support. Best practices: validate with `validate_flow`/`validate_product` before creating, and safely test with `execute_agent`/`run_flow`; report untested paths; for repeatable work, use `create_eval_suite` and `run_eval_suite`; capture real failures with `add_eval_cases`. MCP prompts expose the same guidance; do not fetch both forms for the same task. Report Runtype bugs or doc gaps with `submit_feedback`."}