pash-mcp-server

# Felix Standalone MCP Server **Standalone MCP Server with PASH Compression (Phoenix Architecture)** This project is a minimal, production-ready Model Context Protocol (MCP) server extracted from the Felix ecosystem. It is designed to provide external tools (Cursor, Claude Desktop, etc.) with maximum token savings thanks to built-in PASH compression. ## Features (Phoenix Architecture) - **Pure Core**: No hidden dependencies on the monolithic Felix codebase. - **Dynamic Registration**: Automatic scanning of the `mcp_engines` directory and loading of tools from `manifest.json`. - **PASH Compression**: Intelligent compression of heavy JSON responses (savings >80% tokens) with transparent hash return. - **Zero-config Launch**: Works over `stdio` with configuration via `.env`. ## ⚠️ Safety and Runtime For developers making changes to the server-side (`sse_server.py`, tests): strictly follow the pattern **[Surgical Process Cleanup by Port](docs/RUNTIME_SAFETY.md)**. Global process killing (`taskkill /IM`) is strictly prohibited, as it interrupts the agent session. ## Architecture Boundaries **Felix MCP Standalone** is a self-contained product. It shares **principles, not code** with the internal Felix core (`C:\Users\53\Felix`). - **Product Judge** (in `src/product_judge.py`) is an independent implementation of deterministic validation. It shares the principle of "structural validation before output" with the internal Mirror Forge Synthesis, but has zero shared code by design. This is intentional technical debt prevention. - **Cognitive Loop and PASH compression** are implemented directly in this repository. They do not import from or depend on the internal Felix core. This ensures the product can be deployed, tested, and maintained independently. **Why this matters:** When you install this package, you get a complete, autonomous system. No hidden dependencies on internal infrastructure. ## Installation ```bash git clone https://github.com/AkynZH/pash-mcp-server.git cd pash-mcp-server pip install -e . ``` ## Configuration Copy `.env.example` to `.env` and configure paths: ```env MCP_ENGINES_DIR=./mcp_engines LOG_LEVEL=INFO PASH_MIN_THRESHOLD=500 ``` Create the `mcp_engines` directory and place subfolders with `manifest.json` and executables of your MCP servers. ## Launch ```bash python server.py ``` The server will launch in `stdio` mode and be ready to connect via compatible clients (e.g., Claude Desktop). ## Project Structure ``` pash-mcp-server/ ├── server.py # Entry point (FastMCP) ├── src/ │ ├── config.py # Pydantic configuration │ ├── router.py # Scanner and JSON-RPC proxy │ └── compressor.py # PASH compression core ├── config/ │ └── manifest.example.json # Manifest example └── tests/ # Isolated tests ``` ## Odysseus Integration Status This repository serves as the cognitive core designed to be paired with an interface like Odysseus. - ✅ **Completed & Shipped:** Cognitive Loop, PASH compression, Product Judge, and heuristic routing are fully implemented and tested. - 🗺️ **Roadmap:** Native, one-click UI integration with the Odysseus agent (combining local reflexes with this cognitive core) is currently in active development. *We believe in shipping what works and clearly labeling what is being built.* ## License MIT License (c) 2026 AkynZH