Skip to main content

Tools and Agents

The Model Gateway's Responses API (/v1/responses) is an agentic entry point. Give the model a task and a set of gateway-hosted tools, and the gateway runs the tool-call loop for you: it dispatches each tool call the model makes, feeds the result back, and repeats until the model returns a final answer, with token usage aggregated across the whole run. By default, the gateway returns one buffered response. Set "stream": true to receive the trace as Server-Sent Events while the loop runs. You get multi-step, tool-using behavior from a single API call, with no orchestration to build or host.

Tools over MCP​

Tools are provided through the Model Context Protocol (MCP), from a catalog shared across the platform. The same tools are available to API callers and to the built-in Chat.

  • RAG is available as a tool, so a model can search your documents mid-conversation instead of you calling retrieval separately.
  • Your own MCP servers connect the loop to proprietary systems and data. Deploy an MCP server image — subject to your deployment restrictions — and its tools join the catalog.

Use tools through the API​

Use the Responses API to discover the Model Gateway tools available to your API key, let a model call them, and inspect the complete tool-calling trace.

→ Use Model Gateway tools with the Responses API

Access control​

Every tool invocation is authorized per tenant, just like model and RAG access. Tenants set function tool policy and allowlists, so a request can only reach the tools it's permitted to use.

Client-side tools​

For client-driven agents — coding assistants and similar harnesses — the client runs the agent loop. Local tools such as the filesystem and shell execute on the client, while gateway-hosted RAG and MCP tools are available from the Model Gateway's remote MCP endpoint.

See Configure Coding Agents for an OpenCode setup guide and the settings used by other OpenAI-compatible coding agents.