gcharanteja
feat: implement enhanced 3-tier query search with fuzzy matching and LLM reranking
a964d2d | title: Memory | |
| emoji: ⚡ | |
| colorFrom: indigo | |
| colorTo: purple | |
| sdk: docker | |
| pinned: false | |
| A Docker-powered Model Context Protocol (MCP) server for memory management. | |
| For more details, check out the [Hugging Face Spaces Config Reference](https://huggingface.co). | |
| --- | |
| ## 🛠️ Local Development | |
| Run the server locally with this command: | |
| ```bash | |
| DATA_DIR=. node server.js | |
| ``` | |
| --- | |
| ## ⚙️ Project Configurations | |
| ### 1. `config.json` | |
| Save this file in your root project directory. | |
| ```json | |
| { | |
| "CHROMA_URL": "", | |
| "CHROMA_COLLECTION": "", | |
| "EMBEDDER_BASE_URL": "", | |
| "EMBEDDER_API_KEY": "", | |
| "OPENAI_MODEL": "stepfun-ai/step-3.5-flash", | |
| "OUTPUT_DIR": "/data/context-tree" | |
| } | |
| ``` | |
| ### 2. `mcp.json` (VS Code Integration) | |
| Add this configuration to your local VS Code application path to connect the server. | |
| 📁 **Path:** `~/Library/Application Support/Code/User/mcp.json` | |
| ```json | |
| { | |
| "servers": { | |
| "my-mcp-server-6170c10f": { | |
| "url": "https://maxxcarl-memory.hf.space/sse", | |
| "type": "sse" | |
| } | |
| }, | |
| "inputs": [] | |
| } | |
| ``` | |