> ## Documentation Index
> Fetch the complete documentation index at: https://docs.z.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Tool Integration

## Supported Tools

<Tip>
  The GLM Coding Plan is limited to use within the following officially supported tools and product environments; users may not use their subscription benefits for tools or scenarios outside of this scope.
</Tip>

### 1. Coding Agent Tool

Click on the tool documentation below that you wish to use, and follow the corresponding integration guide to set it up.

<CardGroup cols={3}>
  <Card title="Claude Code" href="/devpack/tool/claude" icon="puzzle-piece">
    The Claude Code IDE plugin supports VSCode and JetBrains.
  </Card>

  <Card title="Claude for IDE" href="/devpack/tool/claude-for-ide" icon="stars">
    The Claude Code IDE plugin supports VSCode and JetBrains.
  </Card>

  <Card title="OpenCode" href="/devpack/tool/opencode" icon="globe">
    The Claude Code IDE plugin supports VSCode and JetBrains.
  </Card>

  <Card title="Cursor" href="/devpack/tool/cursor" icon="code">
    An AI-first code editor that supports custom model configurations.
  </Card>

  <Card title="Cline" href="/devpack/tool/cline" icon="brain">
    An AI programming extension for VS Code that supports code generation and file operations.
  </Card>

  <Card title="TRAE" href="/devpack/tool/trae" icon="headset">
    An AI editor capable of independently completing various development tasks.
  </Card>

  <Card title="Qoder" href="/devpack/tool/qoder" icon="book">
    An agentic coding platform designed for real software development.
  </Card>

  <Card title="Droid" href="/devpack/tool/droid" icon="helmet-safety">
    Enterprise-grade AI coding agent that runs in the terminal to handle end-to-end workflows.
  </Card>

  <Card title="Kilo Code" href="/devpack/tool/kilo" icon="bolt">
    A powerful VS Code extension for code generation and project management.
  </Card>

  <Card title="Roo Code" href="/devpack/tool/roo" icon="box">
    A smart VS Code extension for code writing and refactoring.
  </Card>

  <Card title="Crush" href="/devpack/tool/crush" icon="gauge-high">
    A terminal-based AI programming tool that supports both CLI and TUI interfaces.
  </Card>

  <Card title="Goose" href="/devpack/tool/goose" icon="rocket">
    AI Agent tool, supporting local execution and automated engineering tasks.
  </Card>

  <Card title="Eigent" href="/devpack/tool/goose" icon="rocket">
    A desktop AI agent built on a multi-agent architecture, capable of automating browser, terminal, and MCP-powered workflows.
  </Card>
</CardGroup>

### 2. General-purpose Agent Tool

<Tip>
  The general-purpose agent tools listed below are also supported and will continue to be served on a best-effort basis. Under high inference load (typically around 2–6 PM Singapore time, though this may shift), some requests may face temporary rate limits.
</Tip>

<CardGroup cols={2}>
  <Card title="OpenClaw" href="/devpack/tool/openclaw" icon="lobster">
    An open-source AI assistant that runs on local devices, supports multi-platform use.
  </Card>

  <Card title="Hermes Agent" href="#" icon="shield">
    An open-source evolving AI agent with persistent memory,getting smarter with use.
  </Card>

  <Card title="SillyTavern" href="#" icon="glasses">
    A highly customizable AI chat frontend for immersive roleplay with multi-model & media support.
  </Card>
</CardGroup>

## Coding Endpoint

GLM Coding Plan provides endpoints for two protocols for different coding tools:

`OpenAI Chat Completions` protocol endpoint: `https://api.z.ai/api/coding/paas/v4`

`Anthropic Messages` protocol endpoint: `https://api.z.ai/api/anthropic`

<Warning>
  Please select the correct endpoint address based on the tool you are using. Incorrect endpoint configuration will result in inability to use GLM Coding Plan subscription quota.
</Warning>

## Config Example

> Core Steps:
>
> 1. Select an OpenAI-compatible provider.
> 2. **`Add/Replace the OpenAI Base URL with https://api.z.ai/api/coding/paas/v4.`**
> 3. **`Enter your Z.AI API Key and select GLM-5.1, GLM-5, GLM-5-Turbo, GLM-4.7 or GLM-4.5-air.`**

Using **Cursor** as an example (Note: Custom configuration is only supported in Cursor Pro and higher versions), the following steps demonstrate how to integrate the `GLM-4.7` model via the OpenAI protocol. Similarly, other tools supporting the OpenAI protocol can adopt the same configuration approach.

### 1. Install Cursor

Download and install Cursor from the official website.

### 2. Create a New Provider/Model

In Cursor, navigate to the "**Models**" section and click the "**Add Custom Model**".

![Description](https://cdn.bigmodel.cn/markdown/176032216013820251013-100735.jpeg?attname=20251013-100735.jpeg)

* Select the **OpenAI Protocol**.
* Configure the **OpenAI API Key** (obtained from the Z.AI).
* In **Override OpenAI Base URL**, replace the default URL with `https://api.z.ai/api/coding/paas/v4`.
* Enter the model you wish to use, such as `GLM-5.1`, `GLM-5` or `GLM-4.5-air`.
* Note: In Cursor, the model name must be entered in uppercase, such as `GLM-4.7`.

![Description](https://cdn.bigmodel.cn/markdown/176032218295020251013-100740.jpeg?attname=20251013-100740.jpeg)

### 3. Save and Switch Models

After configuration, save your settings and select the newly created **GLM-4.7 Provider** on the homepage.

### 4. Get Started

With this setup, you can begin using the GLM model for code generation, debugging, task analysis, and more.

![Description](https://cdn.bigmodel.cn/markdown/176032221518820251013-100745.jpeg?attname=20251013-100745.jpeg)

## Step 3: Replacing the API URL

1. **Locate the API configuration section in your tool**:

For example, in **Goose**, this is typically where you set the API address in the configuration file; In **VS Code** plugins or **IntelliJ IDEA** plugins, configuration is usually done through the plugin's settings interface.

2. **Replace the OpenAI Base URL**:

Replace the default OpenAI API URL with `https://api.z.ai/api/coding/paas/v4`.

3. **Enter API Key and Select Model**:

* Enter your **Z.ai API Key**.
* Select `GLM-5.1`(standard, complex tasks) or `GLM-4.5-air`(lightweight, faster response) based on your requirements.
