The Model Context Protocol (MCP) is an open protocol that unleashes the power of LLMs and AI agents into your workflow. MCP acts as a connector between two or more applications. Connect your Zoho Sprints workspace with MCP and then configure your preferred AI agent. You can control the level of access for the AI agent within the platform. Once the configuration is complete, you can quickly fetch project information, generate reports, charts or any kind of data visualization through natural language commands. This saves a lot of manual effort and time for the users.
Pre-requisites
- An AI agent with the appropriate plan to access the integration section.
- A workspace in Zoho Sprints.
- Zoho Sprints API access.
Benefits
- Provide natural language prompts to view or update information
- Generate custom reports and charts quickly
- Fetch all project related activities directly from LLM console
Use case
Let's take the example of a software development team that uses Zoho Sprints to manage agile projects. They can integrate Zoho Sprints with Zoho MCP so AI assistants can securely access sprint data, user stories, bugs, comments, and team activity in real time.
With this integration, you can allow AI to fetch information, modify data, or even delete information from within the AI user interface by providing prompts in natural language. However, ensure that the respective actions are authorized during configuration.
Terminologies
- AI Agents: Software programs that can understand our natural command prompts and perform the requested actions using artificial intelligence.
- MCP Server: It is a backend component that acts as a secure medium between AI agents and external products.
- MCP Client: It the interface where you can enter natural language prompts to trigger the configured actions on the MCP Server.
- MCP Server URL: This URL enables the AI clients to connect with the tools (products) securely.
Connect Zoho MCP and Zoho Sprints
- Login to Zoho MCP.
- Create a MCP server (the server name should only contain only alphanumeric characters and hyphens).
- Add tools (Products) by clicking the Add Tools button.

- Search and select Zoho Sprints.
- Select the actions that you would like the AI to carry out for you.

- Click Add Now.
The actions will be added to the MCP server.

- Now, switch to the Connect tab in the left panel.

The goal of integrating Zoho MCP and Zoho Sprints is to reduce manual work of the developers, sprint owners, and product owners.
Now, connect this Server URL with the MCP Clients.
Connect with Claude
If you choose Claude as your AI agent, you need to add the Server URL in Claude to enable secure access to your Zoho Sprints data.
- Navigate to https://claude.ai/settings/connectors.
- Click Add custom connector.
- Enter the connector name and MCP server URL.
- Click Add.
Once you connect it with Claude, you can create a new chat and ask Claude to fetch or update the required information in Zoho Sprints using natural language.
Note: Provide Workspace ID in the chat window to ensure that the AI client can perform the required action in the workspace.
For example: You can generate a report based on the backlog items available in the workspace.
Connect with Cursor
Cursor AI is the first AI code editor. You can enable AI agent to securely perform real-time actions without leaving your coding workspace.
- Navigate to Connect tab in Zoho MCP.
- In the MCP Clients section, switch to Cursor.

- Click the 'Add to Cursor' button for automatic setup. You can also setup manually by following the steps below:
- Open Cursor and navigate to settings (⇧ + ⌘ + J).
- Go to the MCP Tools tab and click New MCP Server.
- Copy and paste the following JSON configuration below, then save it using CMD + S or CTRL + S.
- Activate Agent Mode.
Make sure the chat is set to Agent mode to access Zoho MCP tools within Cursor. taz
- {
- "mcpServers": {
- "ZohoMCP": {
- "command": "npx",
- "args": [
- "mcp-remote",
- "https://zoho-sprints-server-922466879.zohomcp.com/mcp/9de168bbecc****6318b0738af2a6/message",
- "--transport",
- "http-only"
- ]
- }
- }
- }
Note: MCP server key help connect tools with AI agents. When the key is compromised, anyone with it can access sensitive and confidential information of the connected account. Secure the key safely.
Connect with Windsurf
Windsurf, an advanced AI coding assistant, can be integrated with Zoho Sprints through Zoho MCP. You can easily handle real-time actions right from your coding environment by establishing this connection.
- Open Windsurf settings (⌘ + ,)
- Go to the Cascade tab and click Add Server.
- Select Add custom server +.
Add the following JSON configuration, replacing the placeholder key with your MCP key from below:
Note: Your MCP server URL like a password. Protect it to avoid any unauthorized triggers that can access your sensitive data.
Connect with VS Code
Use AI agent directly from within the Visual Studio code with Zoho MCP. The AI assistant can perform real-time actions securely without having to switch from your coding workspace.
Note: You must have GitHub Copilot enabled and set to Agent mode in Visual Studio Code for Zoho MCP to function correctly.
- Navigate to Connect tab in Zoho MCP.
- In the MCP Clients section, switch to VS Code.
- You can first try clicking the "Add to VS Code" button for automatic setup. Otherwise, you can follow the below mentioned steps to add it manually.
- Open the Visual Studio Code Command Palette (⇧ + ⌘ + P on Mac, Ctrl + Shift + P on Windows)
- Type "MCP: Add Server..." and press Enter.
- Choose "HTTP (HTTP or Server-Sent Events)" and press Enter.
- Paste your MCP key from below into the "Server Key" field and press Enter.
- Provide a server name and press Enter.
- Make sure GitHub Copilot is set to Agent mode.
- Ask GitHub Copilot to use tools from your Zoho MCP server.
Connect with any other MCP Client
If your AI agent, IDE, or automation tool supports the Model Context Protocol, you just need to paste the Server URL.
Transport type
Zoho MCP uses HTTP (Streamable HTTP) transport. Select this if your client asks you to choose: HTTP / Streamable HTTP
JSON config snippet
If your client accepts a JSON config file (common in IDE-based agents), paste this block:
General connection steps
- Open your client's MCP or integrations settings
- Add a new MCP server — look for Add server, New connection, or a mcp.json config file.
- Set transport to HTTP or Streamable HTTP.
- Paste the MCP URL from above into the server URL field.
- Save and enable Agent mode or equivalent in your client.
You can contact our support at
support@zohomcp.com if you have any questions or doubts along the way.