rt-mcp
MCP server for Request Tracker REST2 API, enabling ticket, queue, user, and asset management via natural language.
README
RT MCP Server
Model Context Protocol (MCP) server for Request Tracker REST2 API built with fastmcp.
Features
- ✅ Comprehensive RT REST2 API coverage (85+ tools planned)
- ✅ Tag-based tool filtering (resource type, operation, permission level)
- ✅ Advanced fastmcp features (dependency injection, lifespan, Context)
- ✅ Environment variable configuration
- ✅ Type-safe with Pydantic models
- ✅ Async/await throughout
- ✅ Production-ready error handling
Quick Start
Installation
# Clone or navigate to the repository
cd rt-mcp
# Create virtual environment
python3 -m venv venv
source venv/bin/activate # On Windows: venv\Scripts\activate
# Install dependencies
pip install -e .
# For development
pip install -e ".[dev]"
Configuration
Create a .env file based on .env.example:
cp .env.example .env
Edit .env with your RT credentials:
RT_URL=https://your-rt-server.com
RT_TOKEN=your_token_here
# OR
# RT_USER=username
# RT_PASSWORD=password
Running the Server
The server uses Streamable HTTP transport (listening on port 8000 by default).
# Optional: Configure host/port
export HOST=0.0.0.0
export PORT=8000
# Run directly
python -m rt_mcp.server
# Or use the installed command
rt-mcp
Docker 🐳
# Build and start
docker compose up -d --build
The server will be available at http://localhost:8000/mcp.
Available Tools
Total: 75 tools + 4 resources - Complete RT REST2 API coverage!
Ticket Operations (15 tools, tags: tickets)
create_ticket,get_ticket,update_ticket,delete_ticketsearch_tickets,correspond_ticket,comment_tickettake_ticket,steal_ticket,untake_ticketmerge_tickets,link_ticketsget_ticket_history,get_ticket_attachments
Queue Operations (8 tools, tags: queues)
list_queues,get_queue,create_queue,update_queuesearch_queues,disable_queue,enable_queue
User Operations (11 tools, tags: users)
list_users,get_user,get_current_usercreate_user,update_user,search_usersdisable_user,enable_usergrant_privilege,revoke_privilege
Group Operations (8 tools, tags: groups)
list_groups,get_group,create_group,update_groupdelete_group,search_groupsadd_group_member,remove_group_member
Asset Operations (6 tools, tags: assets)
list_assets,get_asset,create_assetupdate_asset,delete_asset,search_assets
Transaction Operations (3 tools, tags: transactions)
get_transaction,list_transactions,search_transactions
Attachment Operations (3 tools, tags: attachments)
get_attachment,get_attachment_content,upload_attachment
Catalog Operations (6 tools, tags: catalogs)
list_catalogs,get_catalog,create_catalogupdate_catalog,delete_catalog,search_catalogs
Custom Field Operations (6 tools, tags: custom-fields)
list_custom_fields,get_custom_field,create_custom_fieldupdate_custom_field,delete_custom_field,search_custom_fields
Custom Role Operations (6 tools, tags: custom-roles)
list_custom_roles,get_custom_role,create_custom_roleupdate_custom_role,delete_custom_role,search_custom_roles
Advanced Search (3 tools, tags: search)
search_all- Search across all RT objectsbulk_update- Update multiple objects with progress reportingadvanced_ticket_search- Auto-paginated ticket search
Resources (4 resources)
rt://queues/list- List all queuesrt://custom-fields/list- List all custom fieldsrt://user/current- Current authenticated userrt://server/info- RT server information
Tag Taxonomy
Tools are organized with multi-dimensional tags:
Resource Type
tickets- Ticket operationsqueues- Queue managementusers- User operationsgroups- Group managementassets- Asset trackingcatalogs- Catalog managementtransactions- Transaction historyattachments- File operationscustom-fields- Custom field managementcustom-roles- Custom role operations
Operation Type
read- Safe GET operationswrite- State-changing operationsdelete- Destructive operationssearch- Search/query operations
Permission Level
basic- Common user operationspower-user- Advanced operationsadmin- Administrative functions
Using with Claude Desktop
Add to your Claude Desktop configuration (~/Library/Application Support/Claude/claude_desktop_config.json on macOS).
Since the server runs via Streamable HTTP / SSE, configuring it is simple:
{
"mcpServers": {
"rt": {
"url": "http://localhost:8000/mcp"
}
}
}
Note: Ensure the server is running (locally or via Docker) before connecting Claude Desktop.
If you prefer to have Claude Desktop manage the process (local only):
{
"mcpServers": {
"rt": {
"command": "/path/to/rt-mcp/venv/bin/python",
"args": ["-m", "rt_mcp.server"],
"env": {
"RT_URL": "https://your-rt-server.com",
"RT_TOKEN": "your_token_here"
}
}
}
}
Development
Project Structure
src/rt_mcp/
├── __init__.py
├── server.py # FastMCP server with lifespan
├── config.py # Pydantic settings (env vars)
├── client.py # RT REST2 API client (httpx)
├── dependencies.py # Dependency injection providers
├── models/
│ ├── common.py # Base Pydantic models
│ └── ...
├── tools/
│ ├── tickets.py # Ticket tools
│ └── ... # More tool modules coming
├── resources/
│ └── ... # MCP resources
└── utils/
├── errors.py # Exception hierarchy
└── helpers.py
Running Tests
pytest
pytest --cov=rt_mcp
Code Quality
# Format
black src tests
# Lint
ruff check src tests
# Type check
mypy src
Architecture
Dependency Injection
The server uses fastmcp's dependency injection system:
from fastmcp import Context
from fastmcp.dependencies import Depends
from rt_mcp.dependencies import get_rt_client
@mcp.tool()
async def my_tool(
ticket_id: int,
ctx: Context, # Injected MCP context
client = Depends(get_rt_client) # Injected RT client
) -> dict:
await ctx.info("Processing...")
return await client.get_ticket(ticket_id)
Lifespan Management
Server validates RT connection on startup:
@asynccontextmanager
async def lifespan(mcp):
# Startup: validate connection
config = RTConfig()
async with RTClient(config) as client:
await client.validate_connection()
yield
# Shutdown: cleanup
Error Handling
Custom exception hierarchy for RT errors:
RTAuthenticationError(401)RTAuthorizationError(403)RTNotFoundError(404)RTValidationError(422)RTConflictError(409/412)RTNetworkError(network/timeout)RTAPIError(generic)
Roadmap
Phase 1: Foundation ✅
- [x] Project structure
- [x] Configuration with pydantic-settings
- [x] RT client with httpx
- [x] Error handling
- [x] Dependency injection
- [x] Server initialization
Phase 2: Core Tools ✅
- [x] Complete ticket tools (15 total)
- [x] Queue tools (8)
- [x] User tools (11)
- [x] MCP resources (4 resources)
Phase 3: Extended Tools
- [ ] Group tools (8)
- [ ] Asset tools (10)
- [ ] Transaction tools (5)
- [ ] Attachment tools (5)
Phase 4: Advanced Tools
- [ ] Catalog tools (6)
- [ ] Custom field tools (8)
- [ ] Custom role tools (6)
- [ ] Search tools (3)
- [ ] Resources for reference data
Phase 5: Polish
- [ ] Comprehensive tests (>80% coverage)
- [ ] Complete documentation
- [ ] Performance optimization
- [ ] CI/CD pipeline
Contributing
Contributions welcome! Please:
- Fork the repository
- Create a feature branch
- Add tests for new functionality
- Ensure code quality checks pass
- Submit a pull request
License
MIT License
Links
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.