# How to integrate Onedesk MCP with LlamaIndex

```json
{
  "title": "How to integrate Onedesk MCP with LlamaIndex",
  "toolkit": "Onedesk",
  "toolkit_slug": "onedesk",
  "framework": "LlamaIndex",
  "framework_slug": "llama-index",
  "url": "https://composio.dev/toolkits/onedesk/framework/llama-index",
  "markdown_url": "https://composio.dev/toolkits/onedesk/framework/llama-index.md",
  "updated_at": "2026-05-12T10:20:36.219Z"
}
```

## Introduction

This guide walks you through connecting Onedesk to LlamaIndex using the Composio tool router. By the end, you'll have a working Onedesk agent that can log two hours on today's support ticket, remove outdated attachment from project alpha, delete task 'update onboarding guide' from project through natural language commands.
This guide will help you understand how to give your LlamaIndex agent real control over a Onedesk account through Composio's Onedesk MCP server.
Before we dive in, let's take a quick look at the key ideas and tools involved.

## Also integrate Onedesk with

- [OpenAI Agents SDK](https://composio.dev/toolkits/onedesk/framework/open-ai-agents-sdk)
- [Claude Agent SDK](https://composio.dev/toolkits/onedesk/framework/claude-agents-sdk)
- [Claude Code](https://composio.dev/toolkits/onedesk/framework/claude-code)
- [Claude Cowork](https://composio.dev/toolkits/onedesk/framework/claude-cowork)
- [Codex](https://composio.dev/toolkits/onedesk/framework/codex)
- [OpenClaw](https://composio.dev/toolkits/onedesk/framework/openclaw)
- [Hermes](https://composio.dev/toolkits/onedesk/framework/hermes-agent)
- [CLI](https://composio.dev/toolkits/onedesk/framework/cli)
- [Google ADK](https://composio.dev/toolkits/onedesk/framework/google-adk)
- [LangChain](https://composio.dev/toolkits/onedesk/framework/langchain)
- [Vercel AI SDK](https://composio.dev/toolkits/onedesk/framework/ai-sdk)
- [Mastra AI](https://composio.dev/toolkits/onedesk/framework/mastra-ai)
- [CrewAI](https://composio.dev/toolkits/onedesk/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 Onedesk
- Connect LlamaIndex to the Onedesk MCP server
- Build a Onedesk-powered agent using LlamaIndex
- Interact with Onedesk 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 Onedesk MCP server, and what's possible with it?

The Onedesk MCP server is an implementation of the Model Context Protocol that connects your AI agent and assistants like Claude, Cursor, etc directly to your Onedesk account. It provides structured and secure access to your help desk and project management workspace, so your agent can perform actions like managing tickets, handling tasks, logging work, and cleaning up projects on your behalf.
- Automated worklog entry creation: Let your agent log time spent on tickets, tasks, or projects, so you can track team effort without manual entry.
- Ticket and task cleanup: Direct your agent to delete tickets or tasks that are no longer needed, keeping your workspace organized and up to date.
- Project and requirement management: Have the agent remove outdated projects or requirements, ensuring your portfolio stays relevant and clutter-free.
- Attachment and comment removal: Ask your agent to delete attachments or comments from tasks, tickets, or projects, maintaining a clean and focused workflow.
- Customer and timesheet handling: Enable your agent to securely delete customers or timesheets, helping you maintain accurate records and compliance.

## Supported Tools

| Tool slug | Name | Description |
|---|---|---|
| `ONEDESK_CREATE_WORKLOG` | Create Worklog Entry | Create a worklog entry to track time spent on a work item (task, ticket, project, etc.). Records the start and finish time, work hours, completion percentage, and billability. Use this after obtaining a valid item_id from actions like GET_TASK_DETAILS. |
| `ONEDESK_DELETE_ATTACHMENT` | Delete Attachment | Delete a specific attachment from OneDesk by its ID. To use this tool, provide the attachment_id. Note that the OneDesk Public API has limited attachment management capabilities. Attachments are typically associated with items (tickets/tasks) and can be viewed via GET_ATTACHMENTS when fetching item details. Returns deleted=True if the attachment was successfully deleted (200/202/204). Returns deleted=False if the attachment was not found (404). |
| `ONEDESK_DELETE_COMMENT` | Delete Comment | Deletes a comment from OneDesk by its ID. Use this tool to permanently remove a comment when it's no longer needed. Note: OneDesk's Public API has limited DELETE support for comments. This action tries multiple possible endpoints to maximize compatibility. Comments may be referred to as "conversations" in the API. Example: {'comment_id': 12345} Returns: Deletion status with context about success/failure (e.g., comment not found, permission denied) |
| `ONEDESK_DELETE_CUSTOMER` | Delete Customer | Deletes a customer from OneDesk using the DELETE /customers/{id} endpoint. This action permanently removes the customer record from the OneDesk account. Returns: - success=true: Customer was successfully deleted (HTTP 200/202/204) - success=false: Customer not found or already deleted (HTTP 404) Example usage: {'customer_id': '12345'} |
| `ONEDESK_DELETE_MESSAGE` | Delete Message | Deletes a message from OneDesk by its unique message ID. Returns success if the message is deleted or doesn't exist (idempotent operation). Use this when you need to permanently remove a message from the system. |
| `ONEDESK_DELETE_PROJECT` | Delete Project | Delete a project in OneDesk by its ID. **API LIMITATION**: The OneDesk Public API does not support DELETE operations. DELETE requests return 405 Method Not Allowed on /rest/public/items endpoints. The public API only supports: - POST /items/ (create) - GET /items/id/{id} (retrieve by ID) - GET /items/externalId/{externalId} (retrieve by external ID) - GET /organization/profileAndPolicy (organization info) **Alternatives**: - Use OneDesk web application to delete projects manually - Contact OneDesk support for private API access - Use ONEDESK_GET_ISSUE_DETAILS if you need to check item status This action will fail with a clear error explaining the limitation. |
| `ONEDESK_DELETE_REQUIREMENT` | Delete Requirement | Delete a requirement from OneDesk. Requirements in OneDesk are work items used for product development, feature requests, and specifications. They are one of the core work item types (tickets, tasks, requirements, issues). IMPORTANT: This operation may not be supported by all OneDesk Public API versions. If deletion fails, the API may only support soft-delete or archiving through status updates. Use when you need to permanently remove a requirement by its internal numeric ID. Example: {'requirement_id': 12345} |
| `ONEDESK_DELETE_TASK` | Delete Task | Tool to delete a specific task. Use when you need to remove an existing task by its ID after confirming its details. |
| `ONEDESK_DELETE_TICKET` | Delete Ticket | Deletes a ticket from OneDesk by its internal ID. Use this tool to permanently remove a ticket when it's no longer needed. The ticket will be moved to the 'Deleted Items' project and scheduled for permanent deletion after 7 days. Note: OneDesk's Public API has limited DELETE support. This action tries multiple possible endpoints to maximize compatibility. Example: {'ticket_id': 123} Returns: Deletion status with context about success/failure (e.g., ticket not found, rate limited, permission denied) |
| `ONEDESK_DELETE_TIMESHEET` | Delete Timesheet | Tool to delete a timesheet by its ID. Use when you need to remove a specific timesheet after confirming it's no longer needed. Example: {'timesheet_id': 123} |
| `ONEDESK_DELETE_USER` | Delete User | Deletes a user from OneDesk by their unique user ID. This action permanently removes the specified user from the OneDesk account. Use this only after confirming the user should be removed from the system. **Important Notes:** - The OneDesk Public API has limited user management capabilities - There is no public API endpoint to list or create users - User IDs must be obtained through other means (e.g., OneDesk web application, internal systems) **Returns:** - success=True with status_code 200/204 if user was successfully deleted - success=False with status_code 404 if user was not found - success=False with appropriate status_code for other errors (403 for permission denied, 429 for rate limited, etc.) Example: {'user_id': '12345'} |
| `ONEDESK_GET_ATTACHMENTS` | Get Attachments | Retrieve attachments for a specific item by its external ID. IMPORTANT: The OneDesk Public API (/rest/public) has limited functionality and does NOT provide: - A dedicated attachments list endpoint - Ability to list all attachments across items - Endpoints to retrieve items by internal ID This action retrieves attachments by fetching an item via its external ID and extracting attachment data from the item response. You must know the external ID of the item in advance. Supported endpoints: 1. GET /rest/public/items/externalId/{externalId} - Fetch item by external ID 2. POST /rest/public/items/ - Create new items (not used here) 3. GET /rest/public/organization/profileAndPolicy - Get organization info (not used here) |
| `ONEDESK_GET_COMMENT_DETAILS` | Get Comment Details | Retrieve detailed information for a specific comment by its ID. This action attempts to fetch comment details from OneDesk using the comment ID. It tries multiple endpoint patterns to maximize compatibility with different API versions. Note: The OneDesk Public REST API has limited support for retrieving comments. If the comment ID doesn't exist or the endpoint is not available, this action will fail. |
| `ONEDESK_GET_COMMENTS` | Get Comments | Attempts to retrieve comments (discussion posts) from OneDesk. **IMPORTANT LIMITATION**: The OneDesk Public REST API does not provide an endpoint to list or retrieve comments. According to OneDesk's API documentation, the public API only supports: - Creating items: POST /rest/public/items/ - Retrieving items by external ID: GET /rest/public/items/externalId/{externalId} - Getting organization info: GET /rest/public/organization/profileAndPolicy Comments can be created via the API but cannot be retrieved or listed through the public API. This action documents this limitation clearly. **Alternative approaches**: 1. Use the OneDesk web application directly to view comments 2. Request access to OneDesk's private/internal API from OneDesk support 3. Use OneDesk's webhook system to receive comment events in real-time 4. If you have a comment ID, try ONEDESK_GET_COMMENT_DETAILS For more information, visit: https://onedesk.com/dev/ |
| `ONEDESK_GET_FEEDBACK` | Get Feedback Items | Retrieve all feedback items from OneDesk. **IMPORTANT LIMITATION**: The OneDesk Public API does not provide a GET endpoint to list feedback items. The public API only supports: - Creating items via POST /rest/public/items/ - Retrieving specific items by external ID via GET /rest/public/items/externalId/{externalId} - Getting organization info This action will always fail with an informative error message explaining this API limitation. **Alternative approaches:** - Use the OneDesk web application for viewing feedback - Retrieve feedback by external ID if you have it - Use OneDesk webhooks to receive feedback data as events - Contact OneDesk for private/internal API access |
| `ONEDESK_GET_FEEDBACK_DETAILS` | Get Feedback Details | Retrieve detailed information about a specific item in OneDesk by its internal ID. This endpoint retrieves ANY type of item (Tickets, Tasks, Feedback, etc.), not just feedback items. Use this action when you have an item's internal ID and need comprehensive details including: - Basic info (title, description, status, priority) - Creator and assignee information - Project association - Timestamps and lifecycle state - Direct web URL to view the item The internal ID is the numeric ID shown in the OneDesk application. |
| `ONEDESK_GET_ISSUE_DETAILS` | Get Issue Details | Retrieve comprehensive details of a specific issue by its ID. This action fetches full information about an issue including its title, description, status, priority, creation/update timestamps, and other metadata. Use this when you need detailed information about a specific issue that you have the ID for. Returns essential issue information that can be used for reporting, tracking, or further processing. The response includes both basic fields (title, description) and advanced tracking information (status, priority, timestamps). |
| `ONEDESK_GET_ISSUES` | Get Issues | Retrieve a list of issues from OneDesk. **API LIMITATION**: The OneDesk Public API does not provide an endpoint to list or retrieve issues. The public API only supports: 1. Creating items via POST /rest/public/items/ 2. Retrieving specific items by external ID via GET /rest/public/items/externalId/{externalId} 3. Getting organization info via GET /rest/public/organization/profileAndPolicy **Alternatives**: - Use the OneDesk web application directly to view and manage issues - Request access to OneDesk's private API (contact OneDesk support) - Use OneDesk's webhook system for issue event notifications - Use ONEDESK_GET_ISSUE_DETAILS action if you have a specific issue's external ID This action will raise an ExecutionFailed error to document this limitation. |
| `ONEDESK_GET_MESSAGES` | Get Messages | Tool to retrieve a list of messages from OneDesk. **IMPORTANT LIMITATION**: The OneDesk Public API does not provide an endpoint to list or retrieve messages. The public API only supports: 1. Creating items via POST /rest/public/items/ 2. Retrieving specific items by ID via GET /rest/public/items/id/{id} 3. Retrieving specific items by external ID via GET /rest/public/items/externalId/{externalId} 4. Getting organization info via GET /rest/public/organization/profileAndPolicy **Alternatives**: - Use the OneDesk web application directly to view messages - Request access to OneDesk's private API (contact OneDesk support) - Use OneDesk's webhook system for message event notifications - Retrieve specific messages by ID if you have the message ID This action will raise an ExecutionFailed error to document this API limitation. |
| `ONEDESK_GET_REQUIREMENTS` | Get Requirements | Retrieve a list of requirements from OneDesk. **IMPORTANT LIMITATION**: This action cannot function as intended because the OneDesk Public API does not provide an endpoint to list or retrieve requirements. The OneDesk Public API (/rest/public) only supports: - Creating items: POST /rest/public/items/ - Get by external ID: GET /rest/public/items/externalId/{externalId} - Organization info: GET /rest/public/organization/profileAndPolicy There is no endpoint to: - List all requirements - Search for requirements - Query requirements by criteria - Get requirements by any identifier other than external ID Alternative approaches: 1. Use OneDesk web application for requirement management 2. Contact OneDesk support for private API access 3. Use webhooks to receive requirement data as events occur 4. Use ONEDESK_GET_REQUIREMENT_DETAILS if you have a specific requirement's external ID This action will always fail with a clear error message explaining the limitation. |
| `ONEDESK_GET_TASK_DETAILS` | Get Task Details | Retrieves comprehensive details of a specific task/item in OneDesk by its ID. Use this tool when you need complete information about a task including its title, description, status, assignee, priority, due date, and project association. OneDesk uses 'items' to represent various work units including tasks, tickets, issues, and requirements. Returns detailed task information if found, or raises an error if the task doesn't exist or you don't have permission to access it. |
| `ONEDESK_GET_TICKETS` | Get Tickets | Retrieve tickets from OneDesk by querying a range of item IDs. **API LIMITATION WORKAROUND**: The OneDesk Public API does not provide a direct endpoint to list all tickets. This action works around this by: 1. Querying individual items by ID within the specified range (start_id to end_id) 2. Filtering results to return only items with type='Ticket' 3. Limiting results to the max_tickets parameter **How it works**: - The action queries items sequentially from start_id to end_id - Only items with type='Ticket' are included in the results - Stops after collecting max_tickets tickets or reaching end_id - Non-existent IDs and non-ticket items (tasks, folders, etc.) are skipped **Best practices**: - Use smaller ranges (e.g., 1-50) for faster responses - Adjust start_id based on your known ticket ID ranges - Use max_tickets to limit the number of results The OneDesk Public API endpoint used: GET /rest/public/items/id/{id} |
| `ONEDESK_GET_TIMESHEET_DETAILS` | Get Timesheet Details | Tool to retrieve details of a specific timesheet entry. Use when you have the `timesheet_id` and need full metadata (user, project, hours, dates). Tries multiple endpoint/header variants and gracefully falls back in restricted environments. |
| `ONEDESK_GET_WORKLOGS` | Get Worklogs | Retrieve a list of worklogs from OneDesk with optional filtering and pagination. This action attempts to list worklogs by trying multiple candidate endpoints and parameter formats. Note: The OneDesk Public API has limited endpoint support. If no worklogs endpoint is available, this action will return an empty list. Consider using the OneDesk web interface or private API endpoints for full worklog access. Use this action to: - List all worklogs in your OneDesk organization - Filter worklogs by object (ticket/task/project), user, or date range - Paginate through large worklog lists using limit and offset parameters |

## Supported Triggers

None listed.

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

The Onedesk MCP server is an implementation of the Model Context Protocol that connects your AI agent to Onedesk. It provides structured and secure access so your agent can perform Onedesk 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 Onedesk account and project
- Basic familiarity with async Python/Typescript

### 1. Getting API Keys for OpenAI, Composio, and Onedesk

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 Onedesk 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, onedesk)
- 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 Onedesk 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=["onedesk"],
    )

    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 Onedesk actions."
    system_prompt = """
    You are a helpful assistant connected to Composio Tool Router.
    Use the available tools to answer user queries and perform Onedesk 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: ["onedesk"],
    },
  );

  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 Onedesk 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 Onedesk
```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 Onedesk, 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=["onedesk"],
    )

    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 Onedesk actions."
    system_prompt = """
    You are a helpful assistant connected to Composio Tool Router.
    Use the available tools to answer user queries and perform Onedesk 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: ["onedesk"],
    },
  );

  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 Onedesk 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 Onedesk to LlamaIndex through Composio's Tool Router MCP layer.
Key takeaways:
- Tool Router dynamically exposes Onedesk 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 Onedesk MCP Agent with another framework

- [OpenAI Agents SDK](https://composio.dev/toolkits/onedesk/framework/open-ai-agents-sdk)
- [Claude Agent SDK](https://composio.dev/toolkits/onedesk/framework/claude-agents-sdk)
- [Claude Code](https://composio.dev/toolkits/onedesk/framework/claude-code)
- [Claude Cowork](https://composio.dev/toolkits/onedesk/framework/claude-cowork)
- [Codex](https://composio.dev/toolkits/onedesk/framework/codex)
- [OpenClaw](https://composio.dev/toolkits/onedesk/framework/openclaw)
- [Hermes](https://composio.dev/toolkits/onedesk/framework/hermes-agent)
- [CLI](https://composio.dev/toolkits/onedesk/framework/cli)
- [Google ADK](https://composio.dev/toolkits/onedesk/framework/google-adk)
- [LangChain](https://composio.dev/toolkits/onedesk/framework/langchain)
- [Vercel AI SDK](https://composio.dev/toolkits/onedesk/framework/ai-sdk)
- [Mastra AI](https://composio.dev/toolkits/onedesk/framework/mastra-ai)
- [CrewAI](https://composio.dev/toolkits/onedesk/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.
- [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.
- [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.
- [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.
- [Ascora](https://composio.dev/toolkits/ascora) - Ascora is a cloud-based field service management platform for service businesses. It streamlines scheduling, invoicing, and customer operations in one place.
- [Basecamp](https://composio.dev/toolkits/basecamp) - Basecamp is a project management and team collaboration tool by 37signals. It helps teams organize tasks, share files, and communicate efficiently in one place.
- [Beeminder](https://composio.dev/toolkits/beeminder) - Beeminder is an online goal-tracking platform that uses monetary pledges to keep you motivated. Stay accountable and hit your targets with real financial incentives.
- [Boxhero](https://composio.dev/toolkits/boxhero) - Boxhero is a cloud-based inventory management platform for SMBs, offering real-time updates, barcode scanning, and team collaboration. It helps businesses streamline stock tracking and analytics for smarter inventory decisions.
- [Breathe HR](https://composio.dev/toolkits/breathehr) - Breathe HR is cloud-based HR software for SMEs to manage employee data, absences, and performance. It simplifies HR admin, making it easy to keep employee records accurate and up to date.
- [Breeze](https://composio.dev/toolkits/breeze) - Breeze is a project management platform designed to help teams plan, track, and collaborate on projects. It streamlines workflows and keeps everyone on the same page.
- [Bugherd](https://composio.dev/toolkits/bugherd) - Bugherd is a visual feedback and bug tracking tool for websites. It helps teams and clients report website issues directly on live sites for faster fixes.
- [Canny](https://composio.dev/toolkits/canny) - Canny is a platform for managing customer feedback and feature requests. It helps teams prioritize product decisions based on real user insights.
- [Chmeetings](https://composio.dev/toolkits/chmeetings) - Chmeetings is a church management platform for events, members, donations, and volunteers. It streamlines church operations and improves community engagement.

## Frequently Asked Questions

### What are the differences in Tool Router MCP and Onedesk MCP?

With a standalone Onedesk MCP server, the agents and LLMs can only access a fixed set of Onedesk tools tied to that server. However, with the Composio Tool Router, agents can dynamically load tools from Onedesk 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 Onedesk tools.

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

Yes, absolutely. You can configure which Onedesk 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 Onedesk data and credentials are handled as safely as possible.

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