1c

This MCP catalog focuses on 1C:Enterprise MCP servers and related tooling for metadata access, code analysis, and integration flows. The repository presents several frameworks and standalone servers that expose the MCP interface (IDE integrations, metadata services, code analysis, and search). You can run a compatible MCP server via Docker using the image referenced by the mcp_config, then connect AI-assisted tooling or IDE plugins to interact with 1C metadata, run analyses, or fetch structural information about configurations and code. Tools described in the README include metadata explorers, BSL code analysis, graph-based metadata tooling, and RAG-based search across platform documentation and code bases. These servers typically expose HTTP or MCP transports and may offer additional endpoints or streams for runtime interaction and debugging. To use, start the container or deploy via the supported transport, then configure your MCP client (IDE plugin or agent) to point at the running server endpoint and begin issuing Tools/Prompts provided by the server.

Prerequisites:

• Docker installed on your system (Docker Desktop on Windows/macOS, or Docker Engine on Linux)
• Basic familiarity with running containers and exposing ports

Step-by-step:

1. Install Docker if you haven't already.
2. Pull and run the MCP server image (example shown with a placeholder image name used in the repository):

# Pull the image (replace with the correct image name if provided by the project)
docker pull untru-1c-mcp

# Run the container in interactive mode
# This maps port 8080 inside the container to 8080 on the host; adjust as needed
docker run -i -p 8080:8080 untru-1c-mcp

3. Verify the server is up by hitting a basic endpoint (change port/path as appropriate for the image):

curl -i http://localhost:8080/health

4. Configure your MCP client to point at the running server. For the provided Docker setup, the MCP endpoint is typically http://localhost:8080 with the MCP protocol over HTTP or MCP transport, depending on the image features.
5. If you need environment customization, supply the necessary variables via -e flags or a docker-compose file as described in the project docs.

Note: If you use a different deployment method (npx, node, Python, etc.), follow the specific instructions for that transport as described in the corresponding project sections of the README.

Tips and common considerations:

• The repository consolidates multiple MCP frameworks and servers; ensure you run the container image that matches the MCP server you intend to use (metadata, code analysis, or IDE integration).
• Check transport compatibility: some servers expose MCP over custom streams or standard HTTP; align your client accordingly.
• Start with a health endpoint or a simple query to confirm connectivity before attempting heavier tools like graph analyses or RAG searches.
• If you encounter port conflicts, adjust the host port mapping in Docker (for example -p 8081:8080) and update your MCP client endpoint.
• Environment variables (like MCP_LOG_LEVEL or MCP_MODE) are commonly used to tune verbosity and runtime behavior; consult the specific server documentation for supported options.
• When using Docker-based deployment, you may need to supply additional dependencies (e.g., Qdrant, embedding services) if you enable advanced features like semantic search.