Agent frameworks — MCP Server

Use the Raster MCP server from your own agent code — OpenAI Responses, the Agents SDK, LangChain, and the Vercel AI SDK.

Building your own agent? Point it at the Raster MCP server and its tools — list, search, upload, tag, transfer, delete — become available in code.

These run non-interactively, so they authenticate with an organization API key as a Bearer token. Create one and read the full auth model in Authentication.

Detail	Value
URL	`https://mcp.raster.app/`
Transport	Streamable HTTP
Auth	`Authorization: Bearer <API_KEY>`

The Responses API runs the server as a hosted tool. Full options: OpenAI — remote MCP.

from openai import OpenAI

client = OpenAI()
response = client.responses.create(
    model="gpt-5",
    tools=[
        {
            "type": "mcp",
            "server_label": "raster",
            "server_url": "https://mcp.raster.app/",
            "headers": {"Authorization": "Bearer YOUR_RASTER_API_KEY"},
        }
    ],
    input="List my Raster libraries.",
)

The OpenAI Agents SDK adds it as a hosted MCP tool. See Agents SDK — MCP.

from agents import Agent, HostedMCPTool

agent = Agent(
    name="Raster",
    tools=[
        HostedMCPTool(
            tool_config={
                "type": "mcp",
                "server_label": "raster",
                "server_url": "https://mcp.raster.app/",
                "headers": {"Authorization": "Bearer YOUR_RASTER_API_KEY"},
            }
        )
    ],
)

langchain-mcp-adapters turns the server's tools into LangChain / LangGraph tools. See langchain-mcp-adapters.

from langchain_mcp_adapters.client import MultiServerMCPClient

client = MultiServerMCPClient(
    {
        "raster": {
            "transport": "streamable_http",
            "url": "https://mcp.raster.app/",
            "headers": {"Authorization": "Bearer YOUR_RASTER_API_KEY"},
        }
    }
)
tools = await client.get_tools()

The Vercel AI SDK connects over Streamable HTTP. See AI SDK — MCP tools.

import { experimental_createMCPClient } from "ai";
import { StreamableHTTPClientTransport } from "@modelcontextprotocol/sdk/client/streamableHttp.js";

const client = await experimental_createMCPClient({
  transport: new StreamableHTTPClientTransport(new URL("https://mcp.raster.app/"), {
    requestInit: {
      headers: { Authorization: "Bearer YOUR_RASTER_API_KEY" },
    },
  }),
});

const tools = await client.tools();