Gatherings
A server for managing gatherings and sharing expenses.
gatherings MCP Server
A Model Context Protocol server for managing gatherings and expense sharing.
This is a TypeScript-based MCP server that helps track expenses and payments for social events, making it easy to calculate reimbursements and settle balances between friends.
Architecture Note
Important: This project uses a non-standard architecture for a Node.js application. The TypeScript MCP server acts as a wrapper around a Python backend implementation. The Node.js server receives MCP requests, translates them to command-line calls to the Python script, and returns the results.
While this architecture works for our current needs, it's not the most efficient or maintainable approach for a production application.
TODO
- Refactor the implementation to use a more standard architecture:
- Either fully migrate the backend to TypeScript/JavaScript
- Or implement proper inter-process communication between Node.js and Python components
- Or replace with a REST API approach where the Python backend provides an API that the MCP server consumes
- Improve error handling between the Node.js and Python components
- Add automated tests for the integration between components
- Optimise Dockerfile
Features
Tools
create_gathering- Create a new gathering- Takes
gathering_idandmembersas required parameters
- Takes
add_expense- Add an expense for a member- Takes
gathering_id,member_name, andamountas required parameters
- Takes
calculate_reimbursements- Calculate reimbursements for a gathering- Takes
gathering_idas a required parameter
- Takes
record_payment- Record a payment made by a member- Takes
gathering_id,member_name, andamountas required parameters
- Takes
rename_member- Rename an unnamed member- Takes
gathering_id,old_name, andnew_nameas required parameters
- Takes
show_gathering- Show details of a gathering- Takes
gathering_idas a required parameter
- Takes
list_gatherings- List all gatheringsclose_gathering- Close a gathering- Takes
gathering_idas a required parameter
- Takes
delete_gathering- Delete a gathering- Takes
gathering_idas a required parameter, optionalforceparameter
- Takes
add_member- Add a new member to a gathering- Takes
gathering_idandmember_nameas required parameters
- Takes
remove_member- Remove a member from a gathering- Takes
gathering_idandmember_nameas required parameters
- Takes
Development
Install dependencies:
npm install
Build the server:
npm run build
For development with auto-rebuild:
npm run watch
Installation
Configuration
To use with Claude Desktop, add the server config:
{
"mcpServers": {
"gatherings": {
"command": "node",
"args": ["/path/to/gatherings-server/build/index.js"],
"env": {
"GATHERINGS_DB_PATH": "gatherings.db",
"GATHERINGS_SCRIPT": "/path/to/gatherings-server/gatherings.py"
},
"disabled": false,
"autoApprove": [],
"alwaysAllow": [
"create_gathering",
"add_expense",
"calculate_reimbursements",
"record_payment",
"rename_member",
"show_gathering",
"list_gatherings",
"close_gathering",
"delete_gathering",
"add_member",
"remove_member"
],
"timeout": 300
}
}
}
Configuration Options
commandandargs: Specifies how to run the serverenv: Environment variablesGATHERINGS_DB_PATH: Path to the database fileGATHERINGS_SCRIPT: Path to the Python script for handling gathering operations
alwaysAllow: List of tools that will be automatically allowed without promptingtimeout: Maximum execution time in seconds
Debugging
Since MCP servers communicate over stdio, debugging can be challenging. We recommend using the MCP Inspector, which is available as a package script:
npm run inspector
The Inspector will provide a URL to access debugging tools in your browser.
Máy chủ liên quan
Kone.vc
nhà tài trợMonetize your AI agent with contextual product recommendations
URL Shortener
A simple URL shortening tool using the CleanURI API.
Jane
A knowledge management server for stdlib and specs documents, with a configurable storage path.
Context Savvy MCP
Transforms Claude Desktop into a memory-enabled AI assistant with persistent context, secure command execution, and intelligent workflow automation.
Coze Workflow
An MCP server to execute Coze workflows locally via a Python script.
mcp-nfse-nacional
MCP Server to query NFSe on the Brazilian national portal (nfse.gov.br). Search for invoices, download XMLs and PDFs directly from your AI agent.
Notebooklm mcp server
Let your AI agents chat directly with Google NotebookLM for zero-hallucination answers.
Interactive Leetcode MCP
An MCP server enabling guided DSA learning with AI on leetcode.com
ffl-mcp
Give AI a "send file" capability via P2P (Local-first)
FutureSense MCP
556-tool MCP server giving AI agents full access to FutureSense's business platform — invoicing, payroll, accounting, CRM, booking, content creation, website builder, and more.
Sanka
Sanka is the extension company. Plug AI into your sales lifecycle — deals, quotes, billing, and 100+ CRM/commerce integrations — directly from Claude or Codex.