# How to integrate Scale ai MCP with LlamaIndex

```json
{
  "title": "How to integrate Scale ai MCP with LlamaIndex",
  "toolkit": "Scale ai",
  "toolkit_slug": "scale_ai",
  "framework": "LlamaIndex",
  "framework_slug": "llama-index",
  "url": "https://composio.dev/toolkits/scale_ai/framework/llama-index",
  "markdown_url": "https://composio.dev/toolkits/scale_ai/framework/llama-index.md",
  "updated_at": "2026-03-29T06:48:50.832Z"
}
```

## Introduction

This guide walks you through connecting Scale ai to LlamaIndex using the Composio tool router. By the end, you'll have a working Scale ai agent that can create image labeling task for dataset 'road-signs', list completed annotation tasks for project, fetch results of data labeling job through natural language commands.
This guide will help you understand how to give your LlamaIndex agent real control over a Scale ai account through Composio's Scale ai MCP server.
Before we dive in, let's take a quick look at the key ideas and tools involved.

## Also integrate Scale ai with

- [OpenAI Agents SDK](https://composio.dev/toolkits/scale_ai/framework/open-ai-agents-sdk)
- [Claude Agent SDK](https://composio.dev/toolkits/scale_ai/framework/claude-agents-sdk)
- [Claude Code](https://composio.dev/toolkits/scale_ai/framework/claude-code)
- [Claude Cowork](https://composio.dev/toolkits/scale_ai/framework/claude-cowork)
- [Codex](https://composio.dev/toolkits/scale_ai/framework/codex)
- [OpenClaw](https://composio.dev/toolkits/scale_ai/framework/openclaw)
- [Hermes](https://composio.dev/toolkits/scale_ai/framework/hermes-agent)
- [CLI](https://composio.dev/toolkits/scale_ai/framework/cli)
- [Google ADK](https://composio.dev/toolkits/scale_ai/framework/google-adk)
- [LangChain](https://composio.dev/toolkits/scale_ai/framework/langchain)
- [Vercel AI SDK](https://composio.dev/toolkits/scale_ai/framework/ai-sdk)
- [Mastra AI](https://composio.dev/toolkits/scale_ai/framework/mastra-ai)
- [CrewAI](https://composio.dev/toolkits/scale_ai/framework/crew-ai)

## TL;DR

Here's what you'll learn:
- Set your OpenAI and Composio API keys
- Install LlamaIndex and Composio packages
- Create a Composio Tool Router session for Scale ai
- Connect LlamaIndex to the Scale ai MCP server
- Build a Scale ai-powered agent using LlamaIndex
- Interact with Scale ai through natural language

## What is LlamaIndex?

LlamaIndex is a data framework for building LLM applications. It provides tools for connecting LLMs to external data sources and services through agents and tools.
Key features include:
- ReAct Agent: Reasoning and acting pattern for tool-using agents
- MCP Tools: Native support for Model Context Protocol
- Context Management: Maintain conversation context across interactions
- Async Support: Built for async/await patterns

## What is the Scale ai MCP server, and what's possible with it?

The Scale ai MCP server is an implementation of the Model Context Protocol that connects your AI agent and assistants like Claude, Cursor, etc directly to your Scale ai account. It provides structured and secure access so your agent can perform Scale ai operations on your behalf.

## Supported Tools

| Tool slug | Name | Description |
|---|---|---|
| `SCALE_AI_ADD_STUDIO_ASSIGNMENTS` | Add Studio Assignments | Tool to add project assignments to team members in Scale AI Studio. Use when you need to assign specific projects to team members by their email addresses. This action creates new assignments for the specified team members and projects. |
| `SCALE_AI_ADD_TASK_TAGS` | Add Task Tags | Tool to add tags to an existing task. Use when you need to tag or categorize tasks for organization and filtering. Automatically avoids duplicate tags. |
| `SCALE_AI_CREATE_BATCH` | Create Batch | Tool to create a new batch within a project. Use when you need to group multiple tasks together for organizational and processing purposes. |
| `SCALE_AI_CREATE_DOCUMENT_TRANSCRIPTION_TASK` | Create Document Transcription Task | Tool to create a document transcription task where workers transcribe and annotate information from single or multi-page documents. Use when you need to extract structured data from documents like invoices, forms, or screenshots. |
| `SCALE_AI_CREATE_IMAGE_ANNOTATION_TASK` | Create Image Annotation Task | Tool to create an image annotation task where annotators label images with vector geometric shapes (box, polygon, line, point, cuboid, ellipse). Use when you need to annotate objects in images with bounding boxes, polygons, or other geometric annotations. |
| `SCALE_AI_CREATE_LIDAR_ANNOTATION_TASK` | Create Lidar Annotation Task | Tool to create a lidar annotation task where annotators mark objects with 3D cuboids in 3D space. Use when you need to annotate LIDAR frame sequences with 3D object detection. |
| `SCALE_AI_CREATE_LIDAR_SEGMENTATION_TASK` | Create LiDAR Segmentation Task | Tool to create a LiDAR segmentation task where annotators assign semantic class labels to individual LiDAR points. Use when you need to annotate point cloud data with object classes such as vehicles, pedestrians, roads, buildings, etc. Either 'project' or 'batch' must be provided in the request. |
| `SCALE_AI_CREATE_NAMED_ENTITY_RECOGNITION_TASK` | Create Named Entity Recognition Task | Tool to create a named entity recognition task for labelers to highlight text entity mentions. Use when you need to extract and label entities such as people, organizations, or locations from text. |
| `SCALE_AI_CREATE_SEGMENTATION_ANNOTATION_TASK` | Create Segmentation Annotation Task | Tool to create a segmentation task where annotators classify pixels in an image according to provided labels. Use when you need pixel-wise semantic segmentation of images. |
| `SCALE_AI_CREATE_TEXT_COLLECTION_TASK` | Create Text Collection Task | Tool to create a textcollection task for collecting information from attachments and/or web sources. Use when you need to gather structured data from documents, websites, images, or other content by having taskers fill out defined fields. |
| `SCALE_AI_CREATE_VIDEO_ANNOTATION_TASK` | Create Video Annotation Task | Tool to create a video annotation task where annotators draw geometric shapes around specified objects across video frames. Use when you need to annotate video content with bounding boxes, polygons, lines, points, cuboids, or ellipses. Accepts either individual image frames or video files. |
| `SCALE_AI_CREATE_VIDEO_PLAYBACK_ANNOTATION_TASK` | Create Video Playback Annotation Task | Tool to create a video playback annotation task where annotators draw shapes around specified objects in video files. Use when you need to annotate videos with bounding boxes, polygons, lines, points, cuboids, or ellipses for object detection and tracking. |
| `SCALE_AI_DELETE_TASK_TAGS` | Delete Task Tags | Tool to remove specified tags from a Scale AI task. Use when you need to clean up or modify task tags. |
| `SCALE_AI_DELETE_TASK_UNIQUE_ID` | Delete Task Unique ID | Tool to remove the unique identifier from a task. Use when you need to remove a task's unique identifier for enhanced data management control. |
| `SCALE_AI_FINALIZE_BATCH` | Finalize Batch | Tool to finalize a batch so its tasks can be worked on. Use when you need to finalize a batch for Scale Rapid and Studio customers. For other customer types, this endpoint returns success without performing any action. |
| `SCALE_AI_GET_ASSETS` | Get Assets | Tool to retrieve file assets with filtering capabilities by project and metadata. Use when you need to list or search for files uploaded to Scale AI, filtered by project and optionally by metadata. Supports cursor-based pagination for large result sets. |
| `SCALE_AI_GET_BATCH` | Get Batch | Tool to retrieve the details of a batch with the specified name. Use when you need to check the status or configuration of an existing batch. |
| `SCALE_AI_GET_BATCH_STATUS` | Get Batch Status | Tool to retrieve the current status of a batch and task completion counts. Use when you need to monitor batch progress or check how many tasks are pending or completed. |
| `SCALE_AI_GET_FIXLESS_AUDITS` | Get Fixless Audits | Tool to retrieve fixless audits by task ID or audit ID. Use when you need to fetch audit information for quality assessment. At least one of task_id or id must be provided. |
| `SCALE_AI_GET_PROJECT` | Get Project | Tool to retrieve details about a specific Scale AI project using its unique identifier. Use when you need to get project metadata including type, name, parameter history, and creation timestamp. |
| `SCALE_AI_GET_QUALITY_LABELERS` | Get Quality Labelers | Tool to retrieve training attempts matching provided filter parameters. Use when you need to assess labeler performance and understanding of task instructions. At least one of quality_task_ids or labeler_emails must be provided. |
| `SCALE_AI_GET_STUDIO_ASSIGNMENTS` | Get Studio Assignments | Tool to retrieve current project assignments of all active team users in Scale AI Studio. Use when you need to view team member assignments and workload distribution. Excludes invited or disabled team members. |
| `SCALE_AI_GET_STUDIO_BATCHES` | Get Studio Batches | Tool to retrieve basic information about all pending batches in Studio. Use when you need to list batches organized by priority level. |
| `SCALE_AI_GET_TASK` | Get Task | Tool to retrieve detailed information about a specific task in Scale AI. Use when you need to check task status, review task parameters, or access task results. |
| `SCALE_AI_GET_TEAMS` | Get Teams | Tool to retrieve basic information about all team members associated with the account. Use when you need to list team members, check roles, or view notification preferences. |
| `SCALE_AI_GET_TASK_BY_ID` | Get Task by ID | Tool to retrieve detailed information about a specific task using its task ID. Use when you need to check task status, retrieve results, or analyze task metadata. |
| `SCALE_AI_GET_SECURE_TASK_RESPONSE_URL` | Get Secure Task Response URL | Tool to retrieve secure authenticated task response data. Use when you need to access stored response data for 2D segmentation, video, and lidar tasks that cannot be included in the task JSON. |
| `SCALE_AI_IMPORT_FILE` | Import File | Tool to import files from an external URL endpoint into Scale's system rather than uploading directly from local storage. Use when you need to import files from remote URLs for Scale AI projects or data labeling tasks. |
| `SCALE_AI_INVITE_TEAM_MEMBER` | Invite Team Member | Tool to invite users by email to team with specified role. Use when you need to add new team members with roles like labeler, member, or manager. |
| `SCALE_AI_LIST_BATCHES` | List Batches | Tool to retrieve all batches in descending order by creation date. Use when you need to list batches with optional filtering by project, status, or time range. Supports pagination via limit and offset parameters. |
| `SCALE_AI_LIST_PROJECTS` | List Projects | Tool to retrieve information for all projects in the Scale AI account with optional archived filtering. Use when you need to browse or manage project metadata. Returns project details including type, name, parameter history, and creation timestamps. |
| `SCALE_AI_LIST_TASKS` | List Tasks | Tool to retrieve a paginated list of tasks in descending order by creation time. Use when you need to browse tasks with optional filtering by status, type, project, batch, tags, timestamps, or unique identifiers. Supports pagination via limit and next_token. |
| `SCALE_AI_RE_SEND_TASK_CALLBACK` | Re-send Task Callback | Tool to re-send a callback for a completed or errored task to the callback_url. Use when you need to manually trigger a callback resend for a task that has already been processed. |
| `SCALE_AI_REMOVE_STUDIO_ASSIGNMENTS` | Remove Studio Assignments | Tool to unassign projects from specified team members in Scale AI Studio. Use when you need to remove project assignments from one or more team members. |
| `SCALE_AI_RESET_BATCH_PRIORITIES` | Reset Batch Priorities | Tool to restore batch priority order to default order (calibration batches first, then sorted by creation date). Use when you need to reset custom batch priorities back to the default ordering. |
| `SCALE_AI_SET_BATCH_PRIORITIES` | Set Batch Priorities | Tool to modify batch priority order in Scale AI Studio. Use when you need to adjust the priority order of pending batches. You must include all pending studio batches in the request. |
| `SCALE_AI_SET_PROJECT_ONTOLOGY` | Set Project Ontology | Tool to set ontologies on a Scale AI project. Ontologies define the labels or classes that tasks will reference, and projects maintain complete history of ontology versions. Use when you need to configure or update the classification labels for a project. |
| `SCALE_AI_SET_PROJECT_PARAMETERS` | Set Project Parameters | Tool to set default parameters for tasks created under a project. Use when you need to establish or update default parameters that apply to future tasks unless overridden. |
| `SCALE_AI_SET_TASK_METADATA` | Set Task Metadata | Tool to set key-value metadata on an existing Scale AI task. Use when you need to attach custom metadata to track or organize tasks. This operation is idempotent. |
| `SCALE_AI_UPDATE_TASK_UNIQUE_ID` | Update Task Unique ID | Tool to update or assign a unique identifier to a task. Use when you need to set a custom identifier for task tracking in your system. |
| `SCALE_AI_UPLOAD_FILE` | Upload File | Tool to upload a local file to Scale's servers with a maximum size limit of 80 MB per file. Use when you need to upload files for Scale AI projects or data labeling tasks. |

## Supported Triggers

None listed.

## Creating MCP Server - Stand-alone vs Composio SDK

The Scale ai MCP server is an implementation of the Model Context Protocol that connects your AI agent to Scale ai. It provides structured and secure access so your agent can perform Scale ai operations on your behalf through a secure, permission-based interface.
With Composio's managed implementation, you don't have to create your own developer app. For production, if you're building an end product, we recommend using your own credentials. The managed server helps you prototype fast and go from 0-1 faster.

## Step-by-step Guide

### 1. Prerequisites

Before you begin, make sure you have:
- Python 3.8/Node 16 or higher installed
- A Composio account with the API key
- An OpenAI API key
- A Scale ai account and project
- Basic familiarity with async Python/Typescript

### 1. Getting API Keys for OpenAI, Composio, and Scale ai

No description provided.

### 2. Installing dependencies

No description provided.
```python
pip install composio-llamaindex llama-index llama-index-llms-openai llama-index-tools-mcp python-dotenv
```

```typescript
npm install @composio/llamaindex @llamaindex/openai @llamaindex/tools @llamaindex/workflow dotenv
```

### 3. Set environment variables

Create a .env file in your project root:
These credentials will be used to:
- Authenticate with OpenAI's GPT-5 model
- Connect to Composio's Tool Router
- Identify your Composio user session for Scale ai access
```bash
OPENAI_API_KEY=your-openai-api-key
COMPOSIO_API_KEY=your-composio-api-key
COMPOSIO_USER_ID=your-user-id
```

### 4. Import modules

No description provided.
```python
import asyncio
import os
import dotenv

from composio import Composio
from composio_llamaindex import LlamaIndexProvider
from llama_index.core.agent.workflow import ReActAgent
from llama_index.core.workflow import Context
from llama_index.llms.openai import OpenAI
from llama_index.tools.mcp import BasicMCPClient, McpToolSpec

dotenv.load_dotenv()
```

```typescript
import "dotenv/config";
import readline from "node:readline/promises";
import { stdin as input, stdout as output } from "node:process";

import { Composio } from "@composio/core";

import { mcp } from "@llamaindex/tools";
import { agent as createAgent } from "@llamaindex/workflow";
import { openai } from "@llamaindex/openai";

dotenv.config();
```

### 5. Load environment variables and initialize Composio

No description provided.
```python
OPENAI_API_KEY = os.getenv("OPENAI_API_KEY")
COMPOSIO_API_KEY = os.getenv("COMPOSIO_API_KEY")
COMPOSIO_USER_ID = os.getenv("COMPOSIO_USER_ID")

if not OPENAI_API_KEY:
    raise ValueError("OPENAI_API_KEY is not set in the environment")
if not COMPOSIO_API_KEY:
    raise ValueError("COMPOSIO_API_KEY is not set in the environment")
if not COMPOSIO_USER_ID:
    raise ValueError("COMPOSIO_USER_ID is not set in the environment")
```

```typescript
const OPENAI_API_KEY = process.env.OPENAI_API_KEY;
const COMPOSIO_API_KEY = process.env.COMPOSIO_API_KEY;
const COMPOSIO_USER_ID = process.env.COMPOSIO_USER_ID;

if (!OPENAI_API_KEY) throw new Error("OPENAI_API_KEY is not set");
if (!COMPOSIO_API_KEY) throw new Error("COMPOSIO_API_KEY is not set");
if (!COMPOSIO_USER_ID) throw new Error("COMPOSIO_USER_ID is not set");
```

### 6. Create a Tool Router session and build the agent function

What's happening here:
- We create a Composio client using your API key and configure it with the LlamaIndex provider
- We then create a tool router MCP session for your user, specifying the toolkits we want to use (in this case, scale ai)
- The session returns an MCP HTTP endpoint URL that acts as a gateway to all your configured tools
- LlamaIndex will connect to this endpoint to dynamically discover and use the available Scale ai tools.
- The MCP tools are mapped to LlamaIndex-compatible tools and plug them into the Agent.
```python
async def build_agent() -> ReActAgent:
    composio_client = Composio(
        api_key=COMPOSIO_API_KEY,
        provider=LlamaIndexProvider(),
    )

    session = composio_client.create(
        user_id=COMPOSIO_USER_ID,
        toolkits=["scale_ai"],
    )

    mcp_url = session.mcp.url
    print(f"Composio MCP URL: {mcp_url}")

    mcp_client = BasicMCPClient(mcp_url, headers={"x-api-key": COMPOSIO_API_KEY})
    mcp_tool_spec = McpToolSpec(client=mcp_client)
    tools = await mcp_tool_spec.to_tool_list_async()

    llm = OpenAI(model="gpt-5")

    description = "An agent that uses Composio Tool Router MCP tools to perform Scale ai actions."
    system_prompt = """
    You are a helpful assistant connected to Composio Tool Router.
    Use the available tools to answer user queries and perform Scale ai actions.
    """
    return ReActAgent(tools=tools, llm=llm, description=description, system_prompt=system_prompt, verbose=True)
```

```typescript
async function buildAgent() {

  console.log(`Initializing Composio client...${COMPOSIO_USER_ID!}...`);
  console.log(`COMPOSIO_USER_ID: ${COMPOSIO_USER_ID!}...`);

  const composio = new Composio({
    apiKey: COMPOSIO_API_KEY,
    provider: new LlamaindexProvider(),
  });

  const session = await composio.create(
    COMPOSIO_USER_ID!,
    {
      toolkits: ["scale_ai"],
    },
  );

  const mcpUrl = session.mcp.url;
  console.log(`Composio Tool Router MCP URL: ${mcpUrl}`);

  const server = mcp({
    url: mcpUrl,
    clientName: "composio_tool_router_with_llamaindex",
    requestInit: {
      headers: {
        "x-api-key": COMPOSIO_API_KEY!,
      },
    },
    // verbose: true,
  });

  const tools = await server.tools();

  const llm = openai({ apiKey: OPENAI_API_KEY, model: "gpt-5" });

  const agent = createAgent({
    name: "composio_tool_router_with_llamaindex",
        description : "An agent that uses Composio Tool Router MCP tools to perform actions.",
    systemPrompt:
      "You are a helpful assistant connected to Composio Tool Router."+
"Use the available tools to answer user queries and perform Scale ai actions." ,
    llm,
    tools,
  });

  return agent;
}
```

### 7. Create an interactive chat loop

No description provided.
```python
async def chat_loop(agent: ReActAgent) -> None:
    ctx = Context(agent)
    print("Type 'quit', 'exit', or Ctrl+C to stop.")

    while True:
        try:
            user_input = input("\nYou: ").strip()
        except (KeyboardInterrupt, EOFError):
            print("\nBye!")
            break

        if not user_input or user_input.lower() in {"quit", "exit"}:
            print("Bye!")
            break

        try:
            print("Agent: ", end="", flush=True)
            handler = agent.run(user_input, ctx=ctx)

            async for event in handler.stream_events():
                # Stream token-by-token from LLM responses
                if hasattr(event, "delta") and event.delta:
                    print(event.delta, end="", flush=True)
                # Show tool calls as they happen
                elif hasattr(event, "tool_name"):
                    print(f"\n[Using tool: {event.tool_name}]", flush=True)

            # Get final response
            response = await handler
            print()  # Newline after streaming
        except KeyboardInterrupt:
            print("\n[Interrupted]")
            continue
        except Exception as e:
            print(f"\nError: {e}")
```

```typescript
async function chatLoop(agent: ReturnType<typeof createAgent>) {
  const rl = readline.createInterface({ input, output });

  console.log("Type 'quit' or 'exit' to stop.");

  while (true) {
    let userInput: string;

    try {
      userInput = (await rl.question("\nYou: ")).trim();
    } catch {
      console.log("\nAgent: Bye!");
      break;
    }

    if (!userInput) {
      continue;
    }

    const lower = userInput.toLowerCase();
    if (lower === "quit" || lower === "exit") {
      console.log("Agent: Bye!");
      break;
    }

    try {
      process.stdout.write("Agent: ");

      const stream = agent.runStream(userInput);
      let finalResult: any = null;

      for await (const event of stream) {
        // The event.data contains the streamed content
        const data: any = event.data;

        // Check for streaming delta content
        if (data?.delta) {
          process.stdout.write(data.delta);
        }

        // Store final result for fallback
        if (data?.result || data?.message) {
          finalResult = data;
        }
      }

      // If no streaming happened, show the final result
      if (finalResult) {
        const answer =
          finalResult.result ??
          finalResult.message?.content ??
          finalResult.message ??
          "";
        if (answer && typeof answer === "string" && !answer.includes("[object")) {
          process.stdout.write(answer);
        }
      }

      console.log(); // New line after streaming completes
    } catch (err: any) {
      console.error("\nAgent error:", err?.message ?? err);
    }
  }

  rl.close();
}
```

### 8. Define the main entry point

What's happening here:
- We're orchestrating the entire application flow
- The agent gets built with proper error handling
- Then we kick off the interactive chat loop so you can start talking to Scale ai
```python
async def main() -> None:
    agent = await build_agent()
    await chat_loop(agent)

if __name__ == "__main__":
    # Handle Ctrl+C gracefully
    signal.signal(signal.SIGINT, lambda s, f: (print("\nBye!"), exit(0)))
    try:
        asyncio.run(main())
    except KeyboardInterrupt:
        print("\nBye!")
```

```typescript
async function main() {
  try {
    const agent = await buildAgent();
    await chatLoop(agent);
  } catch (err) {
    console.error("Failed to start agent:", err);
    process.exit(1);
  }
}

main();
```

### 9. Run the agent

When prompted, authenticate and authorise your agent with Scale ai, then start asking questions.
```bash
python llamaindex_agent.py
```

```typescript
npx ts-node llamaindex-agent.ts
```

## Complete Code

```python
import asyncio
import os
import signal
import dotenv

from composio import Composio
from composio_llamaindex import LlamaIndexProvider
from llama_index.core.agent.workflow import ReActAgent
from llama_index.core.workflow import Context
from llama_index.llms.openai import OpenAI
from llama_index.tools.mcp import BasicMCPClient, McpToolSpec

dotenv.load_dotenv()

OPENAI_API_KEY = os.getenv("OPENAI_API_KEY")
COMPOSIO_API_KEY = os.getenv("COMPOSIO_API_KEY")
COMPOSIO_USER_ID = os.getenv("COMPOSIO_USER_ID")

if not OPENAI_API_KEY:
    raise ValueError("OPENAI_API_KEY is not set")
if not COMPOSIO_API_KEY:
    raise ValueError("COMPOSIO_API_KEY is not set")
if not COMPOSIO_USER_ID:
    raise ValueError("COMPOSIO_USER_ID is not set")

async def build_agent() -> ReActAgent:
    composio_client = Composio(
        api_key=COMPOSIO_API_KEY,
        provider=LlamaIndexProvider(),
    )

    session = composio_client.create(
        user_id=COMPOSIO_USER_ID,
        toolkits=["scale_ai"],
    )

    mcp_url = session.mcp.url
    print(f"Composio MCP URL: {mcp_url}")

    mcp_client = BasicMCPClient(mcp_url, headers={"x-api-key": COMPOSIO_API_KEY})
    mcp_tool_spec = McpToolSpec(client=mcp_client)
    tools = await mcp_tool_spec.to_tool_list_async()

    llm = OpenAI(model="gpt-5")
    description = "An agent that uses Composio Tool Router MCP tools to perform Scale ai actions."
    system_prompt = """
    You are a helpful assistant connected to Composio Tool Router.
    Use the available tools to answer user queries and perform Scale ai actions.
    """
    return ReActAgent(
        tools=tools,
        llm=llm,
        description=description,
        system_prompt=system_prompt,
        verbose=True,
    );

async def chat_loop(agent: ReActAgent) -> None:
    ctx = Context(agent)
    print("Type 'quit', 'exit', or Ctrl+C to stop.")

    while True:
        try:
            user_input = input("\nYou: ").strip()
        except (KeyboardInterrupt, EOFError):
            print("\nBye!")
            break

        if not user_input or user_input.lower() in {"quit", "exit"}:
            print("Bye!")
            break

        try:
            print("Agent: ", end="", flush=True)
            handler = agent.run(user_input, ctx=ctx)

            async for event in handler.stream_events():
                # Stream token-by-token from LLM responses
                if hasattr(event, "delta") and event.delta:
                    print(event.delta, end="", flush=True)
                # Show tool calls as they happen
                elif hasattr(event, "tool_name"):
                    print(f"\n[Using tool: {event.tool_name}]", flush=True)

            # Get final response
            response = await handler
            print()  # Newline after streaming
        except KeyboardInterrupt:
            print("\n[Interrupted]")
            continue
        except Exception as e:
            print(f"\nError: {e}")

async def main() -> None:
    agent = await build_agent()
    await chat_loop(agent)

if __name__ == "__main__":
    # Handle Ctrl+C gracefully
    signal.signal(signal.SIGINT, lambda s, f: (print("\nBye!"), exit(0)))
    try:
        asyncio.run(main())
    except KeyboardInterrupt:
        print("\nBye!")
```

```typescript
import "dotenv/config";
import readline from "node:readline/promises";
import { stdin as input, stdout as output } from "node:process";

import { Composio } from "@composio/core";
import { LlamaindexProvider } from "@composio/llamaindex";

import { mcp } from "@llamaindex/tools";
import { agent as createAgent } from "@llamaindex/workflow";
import { openai } from "@llamaindex/openai";

dotenv.config();

const OPENAI_API_KEY = process.env.OPENAI_API_KEY;
const COMPOSIO_API_KEY = process.env.COMPOSIO_API_KEY;
const COMPOSIO_USER_ID = process.env.COMPOSIO_USER_ID;

if (!OPENAI_API_KEY) {
    throw new Error("OPENAI_API_KEY is not set in the environment");
  }
if (!COMPOSIO_API_KEY) {
    throw new Error("COMPOSIO_API_KEY is not set in the environment");
  }
if (!COMPOSIO_USER_ID) {
    throw new Error("COMPOSIO_USER_ID is not set in the environment");
  }

async function buildAgent() {

  console.log(`Initializing Composio client...${COMPOSIO_USER_ID!}...`);
  console.log(`COMPOSIO_USER_ID: ${COMPOSIO_USER_ID!}...`);

  const composio = new Composio({
    apiKey: COMPOSIO_API_KEY,
    provider: new LlamaindexProvider(),
  });

  const session = await composio.create(
    COMPOSIO_USER_ID!,
    {
      toolkits: ["scale_ai"],
    },
  );

  const mcpUrl = session.mcp.url;
  console.log(`Composio Tool Router MCP URL: ${mcpUrl}`);

  const server = mcp({
    url: mcpUrl,
    clientName: "composio_tool_router_with_llamaindex",
    requestInit: {
      headers: {
        "x-api-key": COMPOSIO_API_KEY!,
      },
    },
    // verbose: true,
  });

  const tools = await server.tools();

  const llm = openai({ apiKey: OPENAI_API_KEY, model: "gpt-5" });

  const agent = createAgent({
    name: "composio_tool_router_with_llamaindex",
    description:
      "An agent that uses Composio Tool Router MCP tools to perform actions.",
    systemPrompt:
      "You are a helpful assistant connected to Composio Tool Router."+
"Use the available tools to answer user queries and perform Scale ai actions." ,
    llm,
    tools,
  });

  return agent;
}

async function chatLoop(agent: ReturnType<typeof createAgent>) {
  const rl = readline.createInterface({ input, output });

  console.log("Type 'quit' or 'exit' to stop.");

  while (true) {
    let userInput: string;

    try {
      userInput = (await rl.question("\nYou: ")).trim();
    } catch {
      console.log("\nAgent: Bye!");
      break;
    }

    if (!userInput) {
      continue;
    }

    const lower = userInput.toLowerCase();
    if (lower === "quit" || lower === "exit") {
      console.log("Agent: Bye!");
      break;
    }

    try {
      process.stdout.write("Agent: ");

      const stream = agent.runStream(userInput);
      let finalResult: any = null;

      for await (const event of stream) {
        // The event.data contains the streamed content
        const data: any = event.data;

        // Check for streaming delta content
        if (data?.delta) {
          process.stdout.write(data.delta);
        }

        // Store final result for fallback
        if (data?.result || data?.message) {
          finalResult = data;
        }
      }

      // If no streaming happened, show the final result
      if (finalResult) {
        const answer =
          finalResult.result ??
          finalResult.message?.content ??
          finalResult.message ??
          "";
        if (answer && typeof answer === "string" && !answer.includes("[object")) {
          process.stdout.write(answer);
        }
      }

      console.log(); // New line after streaming completes
    } catch (err: any) {
      console.error("\nAgent error:", err?.message ?? err);
    }
  }

  rl.close();
}

async function main() {
  try {
    const agent = await buildAgent();
    await chatLoop(agent);
  } catch (err: any) {
    console.error("Failed to start agent:", err?.message ?? err);
    process.exit(1);
  }
}

main();
```

## Conclusion

You've successfully connected Scale ai to LlamaIndex through Composio's Tool Router MCP layer.
Key takeaways:
- Tool Router dynamically exposes Scale ai tools through an MCP endpoint
- LlamaIndex's ReActAgent handles reasoning and orchestration; Composio handles integrations
- The agent becomes more capable without increasing prompt size
- Async Python provides clean, efficient execution of agent workflows
You can easily extend this to other toolkits like Gmail, Notion, Stripe, GitHub, and more by adding them to the toolkits parameter.

## How to build Scale ai MCP Agent with another framework

- [OpenAI Agents SDK](https://composio.dev/toolkits/scale_ai/framework/open-ai-agents-sdk)
- [Claude Agent SDK](https://composio.dev/toolkits/scale_ai/framework/claude-agents-sdk)
- [Claude Code](https://composio.dev/toolkits/scale_ai/framework/claude-code)
- [Claude Cowork](https://composio.dev/toolkits/scale_ai/framework/claude-cowork)
- [Codex](https://composio.dev/toolkits/scale_ai/framework/codex)
- [OpenClaw](https://composio.dev/toolkits/scale_ai/framework/openclaw)
- [Hermes](https://composio.dev/toolkits/scale_ai/framework/hermes-agent)
- [CLI](https://composio.dev/toolkits/scale_ai/framework/cli)
- [Google ADK](https://composio.dev/toolkits/scale_ai/framework/google-adk)
- [LangChain](https://composio.dev/toolkits/scale_ai/framework/langchain)
- [Vercel AI SDK](https://composio.dev/toolkits/scale_ai/framework/ai-sdk)
- [Mastra AI](https://composio.dev/toolkits/scale_ai/framework/mastra-ai)
- [CrewAI](https://composio.dev/toolkits/scale_ai/framework/crew-ai)

## Related Toolkits

- [Google Sheets](https://composio.dev/toolkits/googlesheets) - Google Sheets is a cloud-based spreadsheet tool for real-time collaboration and data analysis. It lets teams work together from anywhere, updating information instantly.
- [Composio](https://composio.dev/toolkits/composio) - Composio is an integration platform that connects AI agents with hundreds of business tools. It streamlines authentication and lets you trigger actions across services—no custom code needed.
- [Notion](https://composio.dev/toolkits/notion) - Notion is a collaborative workspace for notes, docs, wikis, and tasks. It streamlines team knowledge, project tracking, and workflow customization in one place.
- [Airtable](https://composio.dev/toolkits/airtable) - Airtable combines the flexibility of spreadsheets with the power of a database for easy project and data management. Teams use Airtable to organize, track, and collaborate with custom views and automations.
- [Asana](https://composio.dev/toolkits/asana) - Asana is a collaborative work management platform for teams to organize and track projects. It streamlines teamwork, boosts productivity, and keeps everyone aligned on goals.
- [Google Tasks](https://composio.dev/toolkits/googletasks) - Google Tasks is a to-do list and task management tool integrated into Gmail and Google Calendar. It helps you organize, track, and complete tasks across your Google ecosystem.
- [Linear](https://composio.dev/toolkits/linear) - Linear is a modern issue tracking and project planning tool for fast-moving teams. It helps streamline workflows, organize projects, and boost productivity.
- [Composio search](https://composio.dev/toolkits/composio_search) - Composio search is a unified web search toolkit spanning travel, e-commerce, news, financial markets, images, and more. It lets you and your apps tap into up-to-date web data from a single, easy-to-integrate service.
- [Jira](https://composio.dev/toolkits/jira) - Jira is Atlassian’s platform for bug tracking, issue tracking, and agile project management. It helps teams organize work, prioritize tasks, and deliver projects efficiently.
- [Clickup](https://composio.dev/toolkits/clickup) - ClickUp is an all-in-one productivity platform for managing tasks, docs, goals, and team collaboration. It streamlines project workflows so teams can work smarter and stay organized in one place.
- [Monday](https://composio.dev/toolkits/monday) - Monday.com is a customizable work management platform for project planning and collaboration. It helps teams organize tasks, automate workflows, and track progress in real time.
- [Perplexityai](https://composio.dev/toolkits/perplexityai) - Perplexityai delivers natural, conversational AI models for generating human-like text. Instantly get context-aware, high-quality responses for chat, search, or complex workflows.
- [Browser tool](https://composio.dev/toolkits/browser_tool) - Browser tool is a virtual browser integration that lets AI agents interact with the web programmatically. It enables automated browsing, scraping, and action-taking from any AI workflow.
- [Addressfinder](https://composio.dev/toolkits/addressfinder) - Addressfinder is a data quality platform for verifying addresses, emails, and phone numbers. It helps you ensure accurate customer and contact data every time.
- [Agiled](https://composio.dev/toolkits/agiled) - Agiled is an all-in-one business management platform for CRM, projects, and finance. It helps you streamline workflows, consolidate client data, and manage business processes in one place.
- [Ai ml api](https://composio.dev/toolkits/ai_ml_api) - Ai ml api is a suite of AI/ML models for natural language and image tasks. It provides fast, scalable access to advanced AI capabilities for your apps and workflows.
- [Aivoov](https://composio.dev/toolkits/aivoov) - Aivoov is an AI-powered text-to-speech platform offering 1,000+ voices in over 150 languages. Instantly turn written content into natural, human-like audio for any application.
- [All images ai](https://composio.dev/toolkits/all_images_ai) - All-Images.ai is an AI-powered image generation and management platform. It helps you create, search, and organize images effortlessly with advanced AI capabilities.
- [Anthropic administrator](https://composio.dev/toolkits/anthropic_administrator) - Anthropic administrator is an API for managing Anthropic organizational resources like members, workspaces, and API keys. It helps you automate admin tasks and streamline resource management across your Anthropic organization.
- [Api labz](https://composio.dev/toolkits/api_labz) - Api labz is a platform offering a suite of AI-driven APIs and workflow tools. It helps developers automate tasks and build smarter, more efficient applications.

## Frequently Asked Questions

### What are the differences in Tool Router MCP and Scale ai MCP?

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

### Can I use Tool Router MCP with LlamaIndex?

Yes, you can. LlamaIndex 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 Scale ai tools.

### Can I manage the permissions and scopes for Scale ai while using Tool Router?

Yes, absolutely. You can configure which Scale ai 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.

### How safe is my data with Composio Tool Router?

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 Scale ai data and credentials are handled as safely as possible.

---
[See all toolkits](https://composio.dev/toolkits) · [Composio docs](https://docs.composio.dev/llms.txt)