Leiga MCP Server for Smithery
A Model Context Protocol (MCP) server for the Leiga project management platform, deployed on Smithery.
Overview
This server provides AI agents with the ability to interact with Leiga, a project management tool. It offers comprehensive functionality for searching, creating, and managing issues within Leiga projects.
Features
Tools Available
- search_all_issues - Search for issues across all projects with flexible filtering
- get_issue_detail - Get detailed information about a specific issue
- my_assigned_issues - Get issues assigned to the authenticated user
- create_issue - Create new issues in Leiga projects
- list_project - List all available projects
- current_date - Get the current date
Configuration
The server requires Leiga API credentials to function:
- clientId: Your Leiga API client ID
- secret: Your Leiga API secret
- debug: Optional debug logging flag
Usage
Search Issues
Search for issues using various filters:
- Text search in titles
- Filter by project, status, assignee, priority, sprint, work type
- Date range filtering for start, due, and creation dates
- Configurable result limits
Create Issues
Create new issues with:
- Required: summary and project name
- Optional: description, priority, status, sprint, work type
Issue Management
- Get detailed issue information by ID or issue number
- List projects to find correct project names
- Filter issues assigned to the authenticated user
Authentication
The server uses Leiga's API authentication system. Credentials are configured through Smithery's configuration interface and stored securely.
Development
To run the server locally:
npm run dev
To build for deployment:
npm run build
Deployment
This server is designed to be deployed on the Smithery platform, which handles:
- HTTP transport and routing
- Configuration management
- Authentication and security
- Scaling and availability
API Integration
The server integrates with Leiga's OpenAPI endpoints:
- Issue search and retrieval
- Issue creation and management
- Project listing
- User authentication
Error Handling
The server includes comprehensive error handling for:
- Authentication failures
- API rate limiting
- Invalid parameters
- Network connectivity issues
All errors are returned as user-friendly messages to the AI agent.