io.github.AceDataCloud/mcp-glm
Official# MCP GLM Server
<!-- mcp-name: io.github.AceDataCloud/mcp-glm -->
A Model Context Protocol (MCP) server for Zhipu GLM chat completions via the AceDataCloud platform.
## Features
- **GLM chat completions**: Call Zhipu GLM models through a uniform MCP tool
- **Model discovery**: List the GLM models exposed by AceDataCloud
- **Usage guide**: Inline tool returning the API usage guide
## Installation
```bash
pip install mcp-glm
```
## Configuration
Set your AceDataCloud API token:
```bash
export ACEDATACLOUD_API_TOKEN=your_token_here
```
Get your token from [https://platform.acedata.cloud](https://platform.acedata.cloud).
## Usage
### stdio mode (default)
```bash
mcp-glm
```
### HTTP mode
```bash
mcp-glm --transport http --port 8000
```
## Available Tools
| Tool | Description |
|------|-------------|
| `glm_chat_completions` | Run a GLM chat completion call |
| `glm_list_models` | List available GLM models |
| `glm_get_usage_guide` | Get the API usage guide |
## Documentation
<!-- canonical-documentation -->
[Documentation](https://platform.acedata.cloud/documents/glm-chat-completions)
## License
MIT — see [LICENSE](LICENSE).
TDQS
Scored across 3 tools
Each tool serves a clearly distinct purpose: usage guidance, chat completion generation, and model listing. There is no overlap or ambiguity between them, so an agent would never struggle to choose the right one.
All tool names use the same glm_ prefix and a consistent lower_snake_case style. The verb-noun pattern is uniform (get_usage_guide, chat_completions, list_models), making the set predictable and easy to navigate.
Three tools is well-scoped for a focused GLM chat completions server. Each tool earns its place, and the set feels neither bloated nor too thin.
The server covers the core lifecycle of chat completion: listing models and making completions, plus an onboarding guide. It lacks endpoints for embeddings or model-specific detail, but those are likely outside the intended scope.