postgsail-mcp-server

Model Context Protocol (MCP) server tailored for PostgSail. that provides AI agents with read-only access to PostgSail marine data systems. This server enables LeChat, Claude and other AI assistants to search and navigate logs, moorages, and stays, monitor and analyze your boat all in one place.

https://github.com/user-attachments/assets/162dac63-8f80-4e46-9357-1033b6b6c78b

Remote MCP Server

Accessible at https://mcp.openplotter.cloud/

LeChat allow remote connectors even with a free account.

Others AI ClaudeAI and OpenAI requires an upgrade plan to use remote connectors.

However you can use the MCP locally with Claude Desktop app.

Installation

• Install Claude App
• Settings -> Extensions
• Drag .MCPB or .DXT files here to install

OR

• Install Claude App.
• Settings -> Extensions -> Advanced settings -> Extension Developer
• Install Extension...

Key Features

• Daily vessel summaries (current status, weather, systems)
• Voyage planning (reviewing past trips, moorage information)
• System monitoring (battery, solar, sensors, connectivity)
• Historical analysis (tracking patterns, favorite destinations)
• Data export (for external navigation tools)
• Maintenance tracking (through stay notes and logs)

Examples

You can ask anything related to PostgSail all readonly API endpoint are available, eg: Logbook, Moorages, Stays, Monitoring. A few examples:

• Provide a daily briefing of <boat_name> systems
• Summarize my last trip
• Summarize my summer sailing voyage

Available Resources

The server exposes these static reference resources:

• postgsail://postgsail_overview - PostgSail overview and core concepts
• postgsail://path_categories_guide - Guide to understanding PostgSail paths
• postgsail://mcp_tool_reference - Reference for available MCP tools and usage patterns

Note: While these resources can be accessed individually via the MCP resource protocol, the get_initial_context() tool provides a more convenient way to access all reference materials in a single call, making it the recommended approach for AI agents to understand the system.

License

Apache License - see LICENSE file for details.

Contributing

1. Fork the repository
2. Create a feature branch
3. Make your changes with tests
4. Run npm run test and npm run typecheck
5. Submit a pull request

Support

For issues and questions:

• Check the troubleshooting section above
• Review SignalK server logs and configuration
• Open an issue with connection details and error messages