==================
/// MCP ///
/// MCP ///
==================
[server:online]
[protocol:ready]mcp-hfspace
by evalstate
Node-based MCP Server that lets you expose one or more Hugging Face Spaces as Model Context Protocol (MCP) endpoints, with special integration for Claude Desktop (file handling, desktop-mode, tool discovery, etc.).
330
48
Open SourceInstallation
1. Prerequisites
• Node.js ≥ 18 and npm (or pnpm/yarn)
• A HuggingFace access-token with “inference” scope
• (Optional) Claude API key if you intend to use Claude-Desktop mode
2. Clone the repository
git clone https://github.com/evalstate/mcp-hfspace.git
cd mcp-hfspace
3. Install dependencies
npm install # or pnpm install | yarn
4. Build the TypeScript sources
npm run build # emits JS to ./dist
5. Configure environment
Create a .env file or export variables in your shell:
HF_TOKEN="<your_huggingface_token>" CLAUDE_TOKEN="<your_claude_key>" # optional PORT=3333 # default port the MCP server listens on6. Start the server
npm start # or: node dist/index.js
7. Verify
Open http://localhost:3333/health to check the health probe.
Docker (alternative):
docker build -t mcp-hfspace . docker run -e HF_TOKEN=xxxxx -p 3333:3333 mcp-hfspaceDocumentation
# Transform Augment Code into a Multimodal AI Development Powerhouse
**mcp-hfspace** turns Augment Code into your ultimate development companion by unlocking access to hundreds of HuggingFace Spaces - from FLUX image generation to Qwen language models to Whisper speech-to-text. Instead of juggling multiple AI tools, you can now have Augment handle everything from code generation to asset creation, all within your development workflow.
## Seamless Integration with Your Development Workflow
Adding this MCP server to your Augment Code setup is straightforward - just configure it in your MCP settings and specify which HuggingFace Spaces you want access to. The server automatically discovers the best endpoints and configures them for immediate use. With intelligent file handling and working directory management, Augment can now generate images for your UI mockups, create audio assets for your applications, analyze screenshots of interfaces, and transcribe audio files - all while you're coding.
## Real-World Development Scenarios
Imagine asking Augment to "generate a hero image for this landing page using FLUX, then analyze this screenshot with OmniParser to extract the layout structure and generate the corresponding React components." Or having it transcribe user feedback audio with Whisper, then use Qwen for intelligent code reviews and suggestions. The server supports everything from background removal for design assets to text-to-speech for accessibility features, making Augment your one-stop solution for modern multimodal development tasks.
**Quick Setup Example:**
```json
{
"mcpServers": {
"mcp-hfspace": {
"command": "npx",
"args": [
"-y",
"@llmindset/mcp-hfspace",
"--work-dir=/your/project/assets",
"black-forest-labs/FLUX.1-schnell",
"microsoft/OmniParser",
"hf-audio/whisper-large-v3-turbo",
"Qwen/Qwen2.5-72B-Instruct"
]
}
}
}
```
With this setup, Augment Code becomes a comprehensive AI development assistant that can handle visual design, audio processing, advanced reasoning, and code generation - all integrated into your existing workflow without switching tools.License: MIT License
Updated 7/15/2025