How to integrate VLM Run MCP with Claude Code

Manage your VLM Run directly from Claude Code with zero worries about OAuth hassles, API-breaking issues, or reliability and security concerns. You can do this in two different ways: Via Composio Connect - Direct and easiest approach Via Composio SDK - Programmatic approach with more control

VLM Run logoVLM Run
Api Key

VLM Run is a multimodal AI platform for structured extraction, predictions, files, skills, feedback, and evaluations. It helps teams build reliable vision-language workflows without stitching together model, file, and eval infrastructure by hand.

11 Tools

Introduction

Manage your VLM Run directly from Claude Code with zero worries about OAuth hassles, API-breaking issues, or reliability and security concerns.

You can do this in two different ways:

  1. Via Composio Connect - Direct and easiest approach
  2. Via Composio SDK - Programmatic approach with more control

Also integrate VLM Run with

Why use Composio?

  • Only one MCP URL to connect multiple apps with Claude Code with zero auth hassles.
  • Programmatic tool calling allows LLMs to write its code in a remote workbench to handle complex tool chaining. Reduces to-and-fro with LLMs for frequent tool calling.
  • Handling Large tool responses out of LLM context to minimize context rot.
  • Dynamic just-in-time access to 20,000 tools across 1000+ other Apps for cross-app workflows. It loads the tools you need, so LLMs aren't overwhelmed by tools you don't need.

Connecting VLM Run to Claude Code using Composio

1. Add the Composio MCP to Claude

Terminal

2. Start Claude Code

bash
claude

3. Open your MCP list

bash
/mcp

4. Select Composio and click on Authenticate

Select Composio and click Authenticate

5. This will redirect you to the Composio OAuth page. Complete the flow by authorizing Composio and you're all set.

Composio OAuth authorization page
Composio authorization complete
Ask Claude to connect to your account and authenticate via the link

What is the VLM Run MCP server, and what's possible with it?

The VLM Run MCP server is an implementation of the Model Context Protocol that connects your AI agent and assistants like Claude, Cursor, etc directly to your VLM Run account. It provides structured and secure access to your multimodal agents, files, skills, extraction workflows, and runs, so your agent can upload files, extract structured data, execute agents, create reusable skills, and track results on your behalf.

  • Structured data extraction: Have your agent extract structured JSON from documents, images, videos, or audio using a supported domain, custom schema, or existing skill.
  • Multimodal agent execution: Direct your agent to choose an available VLM Run agent or provide an inline configuration, then run it against multimodal inputs.
  • File and skill management: Let the agent upload local files, find previously uploaded files, browse existing skills, or create a reusable skill from an uploaded zip, prompt, or chat session.
  • Run tracking and result retrieval: Instruct your agent to list extraction and agent runs, check work in progress, and retrieve completed results.
  • Schema and artifact discovery: Have your agent explore supported extraction domains and schemas, then list artifacts created during a chat session or agent execution.

Connecting VLM Run via Composio SDK

Composio SDK is the underlying tech that powers Rube. It's a universal gateway that does everything Rube does but with much more programmatic control. You can programmatically generate an MCP URL with the app you need (here VLM Run) for even more tool search precision. It's secure and reliable.

How the Composio SDK works

The Composio SDK follows a three-phase workflow:

  1. Discovery: Searches for tools matching your task and returns relevant toolkits with their details.
  2. Authentication: Checks for active connections. If missing, creates an auth config and returns a connection URL via Auth Link.
  3. Execution: Executes the action using the authenticated connection.

Step-by-step Guide

Step by step10 STEPS
1

Prerequisites

Before starting, make sure you have:
  • Claude Pro, Max, or API billing enabled Anthropic account
  • Composio API Key
  • A VLM Run account
  • Basic knowledge of Python or TypeScript
2

Install Claude Code

bash
# macOS, Linux, WSL
curl -fsSL https://claude.ai/install.sh | bash

# Windows PowerShell
irm https://claude.ai/install.ps1 | iex

# Windows CMD
curl -fsSL https://claude.ai/install.cmd -o install.cmd && install.cmd && del install.cmd

To install Claude Code, use one of the following methods based on your operating system:

3

Set up Claude Code

bash
cd your-project-folder
claude

Open a terminal, go to your project folder, and start Claude Code:

  • Claude Code will open in your terminal
  • Follow the prompts to sign in with your Anthropic account
  • Complete the authentication flow
  • Once authenticated, you can start using Claude Code
Claude Code initial setup showing sign-in prompt
Claude Code terminal after successful login
4

Set up environment variables

bash
COMPOSIO_API_KEY=your_composio_api_key_here
USER_ID=your_user_id_here

Create a .env file in your project root with the following variables:

  • COMPOSIO_API_KEY authenticates with Composio (get it from Composio dashboard)
  • USER_ID identifies the user for session management (use any unique identifier)
5

Install Composio library

npm install @composio/core dotenv

Install the Composio TypeScript library to create MCP sessions.

  • @composio/core provides the core Composio functionality
  • dotenv loads environment variables from your .env file
6

Generate Composio MCP URL

import 'dotenv/config';
import { Composio } from '@composio/core';

const { COMPOSIO_API_KEY, USER_ID } = process.env;

if (!COMPOSIO_API_KEY || !USER_ID) {
  throw new Error('COMPOSIO_API_KEY and USER_ID required in .env');
}

const composioClient = new Composio({ apiKey: COMPOSIO_API_KEY });

const composioSession = await composioClient.create(USER_ID, {
  toolkits: ['vlm_run'],
});

const composioMcpUrl = composioSession?.mcp.url;

console.log(`MCP URL: ${composioMcpUrl}`);
console.log(`\nUse this command to add to Claude Code:`);
console.log(`claude mcp add --transport http vlm_run-composio "${composioMcpUrl}" --headers "X-API-Key:${COMPOSIO_API_KEY}"`);

Create a script to generate a Composio MCP URL for VLM Run. This URL will be used to connect Claude Code to VLM Run.

What's happening

  • We import the Composio client and load environment variables
  • Create a Composio instance with your API key
  • Call create() to create a Tool Router session for VLM Run
  • The returned mcp.url is the MCP server URL that Claude Code will use
  • The script prints this URL so you can copy it
7

Run the script and copy the MCP URL

node --loader ts-node/esm generate_mcp_url.ts
# or if using tsx
tsx generate_mcp_url.ts

Run your TypeScript script to generate the MCP URL.

  • The script connects to Composio and creates a Tool Router session
  • It prints the MCP URL and the exact command you need to run
  • Copy the entire claude mcp add command from the output
8

Add VLM Run MCP to Claude Code

bash
claude mcp add --transport http vlm_run-composio "YOUR_MCP_URL_HERE" --headers "X-API-Key:YOUR_COMPOSIO_API_KEY"

# Then restart Claude Code
exit
claude

In your terminal, add the MCP server using the command from the previous step. The command format is:

  • claude mcp add registers a new MCP server with Claude Code
  • --transport http specifies that this is an HTTP-based MCP server
  • The server name (vlm_run-composio) is how you'll reference it
  • The URL points to your Composio Tool Router session
  • --headers includes your Composio API key for authentication

After running the command, close the current Claude Code session and start a new one for the changes to take effect.

9

Verify the installation

bash
claude mcp list

Check that your VLM Run MCP server is properly configured.

  • This command lists all MCP servers registered with Claude Code
  • You should see your vlm_run-composio entry in the list
  • This confirms that Claude Code can now access VLM Run tools

If everything is wired up, you should see your vlm_run-composio entry listed:

Claude Code MCP list showing the toolkit MCP server
10

Authenticate VLM Run

The first time you try to use VLM Run tools, you'll be prompted to authenticate.

  • Claude Code will detect that you need to authenticate with VLM Run
  • It will show you an authentication link
  • Open the link in your browser (or copy/paste it)
  • Complete the VLM Run authorization flow
  • Return to the terminal and start using VLM Run through Claude Code

Once authenticated, you can ask Claude Code to perform VLM Run operations in natural language. For example:

  • "Extract invoice fields from uploaded PDFs"
  • "Classify product images by defect type"
  • "Evaluate extraction quality against gold labels"

Complete Code

Here's the complete code to get you started with VLM Run and Claude Code:

import 'dotenv/config';
import { Composio } from '@composio/core';

const { COMPOSIO_API_KEY, USER_ID } = process.env;

if (!COMPOSIO_API_KEY || !USER_ID) {
  throw new Error('COMPOSIO_API_KEY and USER_ID required in .env');
}

const composioClient = new Composio({ apiKey: COMPOSIO_API_KEY });

const composioSession = await composioClient.create(USER_ID, {
  toolkits: ['vlm_run'],
});

const composioMcpUrl = composioSession?.mcp.url;

console.log(`MCP URL: ${composioMcpUrl}`);
console.log(`\nUse this command to add to Claude Code:`);
console.log(`claude mcp add --transport http vlm_run-composio "${composioMcpUrl}" --headers "X-API-Key:${COMPOSIO_API_KEY}"`);

Conclusion

You've successfully integrated VLM Run with Claude Code using Composio's MCP server. Now you can interact with VLM Run directly from your terminal using natural language commands.

Key features of this setup:

  • Terminal-native experience without switching contexts
  • Natural language commands for VLM Run operations
  • Secure authentication through Composio's managed MCP
  • Tool Router for dynamic tool discovery and execution

Next steps:

  • Try asking Claude Code to perform various VLM Run operations
  • Add more toolkits to your Tool Router session for multi-app workflows
  • Integrate this setup into your development workflow for increased productivity

You can extend this by adding more toolkits, implementing custom workflows, or building automation scripts that leverage Claude Code's capabilities.

TOOLS

Supported Tools

Every VLM Run action and event your agent gets out of the box.

Create Skill

Create a reusable skill from exactly one uploaded zip, prompt, or chat session.

Discover Extraction Schemas

List supported structured-extraction domains, or return the full JSON schema for one domain when domain is provided.

Execute Agent

Start a VLM Run agent execution from an existing agent name or inline configuration over multimodal inputs.

Extract Structured JSON

Start structured JSON extraction from images, a document, a video, or audio using a domain, custom schema, or skill.

Find Files

List uploaded files or find one by file ID or MD5 hash.

Find Skills

List VLM Run skills or find one exact skill by ID, name, and optional version.

Get Run

Get the current status and result of one structured-extraction prediction or agent execution; call repeatedly to poll asynchronous work.

List Agents

Return agents available to the connected account for selection before execution.

List Artifacts

List artifact metadata belonging to exactly one chat session or agent execution.

List Runs

List structured-extraction predictions or agent executions for the connected account.

Upload File

Upload a local file to VLM Run for extraction, agent input, or skill creation.

FAQ

Frequently asked questions

With a standalone VLM Run MCP server, the agents and LLMs can only access a fixed set of VLM Run tools tied to that server. However, with the Composio Tool Router, agents can dynamically load tools from VLM Run and many other apps based on the task at hand, all through a single MCP endpoint.

Yes, you can. Claude Code fully supports MCP integration. You get structured tool calling, message history handling, and model orchestration while Tool Router takes care of discovering and serving the right VLM Run tools.

Yes, absolutely. You can configure which VLM Run scopes and actions are allowed when connecting your account to Composio. You can also bring your own OAuth credentials or API configuration so you keep full control over what the agent can do.

All sensitive data such as tokens, keys, and configuration is fully encrypted at rest and in transit. Composio is SOC 2 Type 2 compliant and follows strict security practices so your VLM Run data and credentials are handled as safely as possible.

Start with VLM Run.It takes 30 seconds.

Managed auth, hosted MCP servers, and every VLM Run tool your agent needs.Free to start.

Start building