MCP

Croma exposes every data source as a tool over the Model Context Protocol (MCP), served over Streamable HTTP at:

https://api.croma.run/mcp

Any MCP client can connect. Interactive clients (Claude, ChatGPT, Cursor) use the OAuth flow: add Croma to Claude in one click and sign in with your Croma account. For your own code you authenticate with the same API key as the REST API: send it as a bearer token and the tools run scoped to your organization, sharing the same rate limits and usage.

Use it from the AI SDK

The AI SDK can load the Croma tools directly and hand them to a model for tool calling. Pass your key in the Authorization header of the transport:

import { experimental_createMCPClient as createMCPClient, generateText } from "ai";
import { StreamableHTTPClientTransport } from "@modelcontextprotocol/sdk/client/streamableHttp.js";
import { anthropic } from "@ai-sdk/anthropic";

const mcp = await createMCPClient({
  transport: new StreamableHTTPClientTransport(
    new URL("https://api.croma.run/mcp"),
    {
      requestInit: {
        headers: { Authorization: `Bearer ${process.env.CROMA_API_KEY}` },
      },
    },
  ),
});

// Every data source, exposed as a tool the model can call.
const tools = await mcp.tools();

const { text } = await generateText({
  model: anthropic("claude-opus-4-8"),
  tools,
  // Let the model call tools and then answer. Consult the AI SDK docs for the
  // multi-step setting in your version (`stopWhen` / `maxSteps`).
  prompt: "Consulta los antecedentes de la Policía Nacional para la cédula 1234567890.",
});

await mcp.close();
console.log(text);

Treat the key as a secret: load it from an environment variable or secrets manager, never commit it, and close the client (mcp.close()) when you are done so the connection is released.

Start from the chat template

Croma Chat Template is an open-source Next.js chat built on this endpoint with the AI SDK and AI Elements: streaming responses, tool calls across every Croma source, a per-message source picker, and optional rate limiting. Try the live deployment at chat.usecroma.com, or use the repository’s one-click Vercel deploy: it asks for your CROMA_API_KEY and a model key and comes up ready to use.

Tool names

Tools are named after their source, with underscores instead of hyphens, for example policia_criminal_records, rues_entity_by_nit, and rama_judicial_cases_by_radicado. Call mcp.tools() to list the full set with their input schemas, or browse the API reference for the fields each one accepts. Async lookups always wait inline over MCP: the tool call returns the finished result, with no 202, polling, or callbacks. The monitor tools manage monitors for the connected organization: create_monitor watches a dataset endpoint on a schedule and emails what is new, list_monitors, get_monitor, update_monitor (rename, re-query, pause, resume), run_monitor, list_monitor_matches and delete_monitor do the rest. They are the only tools that write. Monitors from Claude has prompts that work and what the agent does with them.

Any MCP client

The AI SDK is one option. Because the endpoint speaks standard MCP over Streamable HTTP, the official MCP TypeScript SDK, LangChain, and other MCP-aware frameworks connect the same way: point them at https://api.croma.run/mcp and set the Authorization: Bearer header.

Get Started

Request types

Monitors

Countries

Use it from the AI SDK

Start from the chat template

Tool names

Any MCP client

​Use it from the AI SDK

​Start from the chat template

​Tool names

​Any MCP client

Use it from the AI SDK

Start from the chat template

Tool names

Any MCP client