- Home
- MCP servers
- Lark OpenAPI
Lark OpenAPI
- typescript
406
GitHub Stars
typescript
Language
4 months ago
First Indexed
3 weeks ago
Catalog Refreshed
Documentation & install
Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.
Installation
Add the following to your MCP client configuration file.
Configuration
View docs{
"mcpServers": {
"larksuite-lark-openapi-mcp": {
"command": "npx",
"args": [
"-y",
"@larksuiteoapi/lark-mcp",
"mcp",
"-a",
"<your_app_id>",
"-s",
"<your_app_secret>"
]
}
}
}You connect Feishu/Lark Open Platform APIs to your AI workflows using the Feishu/Lark MCP server. It exposes the platform interfaces as MCP tools, enabling automation tasks like document processing, calendar events, conversations, and more by letting AI agents call these APIs directly through a simple, configurable client.
How to use
Set up a local MCP client that talks to the Feishu/Lark MCP server, then enable API calls from your AI workflows. You will configure a stdio MCP server, provide your app credentials, and start the MCP client to access Feishu/Lark APIs either with a system-wide application or with user authentication when required.
How to install
Prerequisites you need on your machine: Node.js (LTS) and npm. You will run the MCP client with npx to connect to the Feishu/Lark MCP server.
Follow these steps to prepare and run the MCP client.
Step 1: Prepare your Feishu/Lark app credentials
Create a Feishu/Lark application and obtain your App ID and App Secret. You will use these values to authenticate API calls via the MCP client.
Step 2: Configure the MCP client
Add the MCP server configuration to your local configuration so the client can start the MCP service and route API calls to Feishu/Lark.
{
"mcpServers": {
"lark_mcp": {
"command": "npx",
"args": [
"-y",
"@larksuiteoapi/lark-mcp",
"mcp",
"-a",
"<your_app_id>",
"-s",
"<your_app_secret>"
]
}
}
}
Step 3: Run the MCP client
Start the MCP client using the configuration you created. The command listed in the configuration will launch the MCP server and expose the Feishu/Lark API calls to your AI tools.
Step 4: Optional user login for user-scoped APIs
If you need to access APIs with user identity, log in to obtain a user access token. You must configure the OAuth redirect URL in the developer console before using the login flow.
npx -y @larksuiteoapi/lark-mcp login -a cli_xxxx -s yyyyy
Step 5: Optional OAuth token mode
If you want APIs to be called with user access tokens, update the MCP configuration to enable OAuth and set the token mode to user_access_token.
{
"mcpServers": {
"lark_mcp": {
"command": "npx",
"args": [
"-y",
"@larksuiteoapi/lark-mcp",
"mcp",
"-a",
"<your_app_id>",
"-s",
"<your_app_secret>",
"--oauth",
"--token-mode", "user_access_token"
]
}
}
}
Step 6: Domain configuration
Choose Feishu (China) or Lark (International) domain to target the appropriate environment. Set the domain in the MCP client configuration if you are using the international version.
{
"mcpServers": {
"lark_mcp": {
"command": "npx",
"args": [
"-y",
"@larksuiteoapi/lark-mcp",
"mcp",
"-a",
"<your_app_id>",
"-s",
"<your_app_secret>",
"--domain",
"https://open.larksuite.com"
]
}
}
}
Additional notes on usage
Non-preset APIs may not be fully tested for compatibility. Use the preset tool collections to start with well-supported APIs and expand as needed.
Security and access
Keep your App Secret secure. Do not publish credentials in public places. If you enable user authentication, ensure your redirect URL and scopes align with your security requirements.
Troubleshooting and tips
If you encounter permission issues when calling user resources, verify that you are using the correct token mode and that the domain configuration matches your app’s environment.
Development and examples
You can find sample integration patterns showing how MCP enables triggering tool calls and messaging through Feishu/Lark bots and conversations.