Yamcs MCP Server
Enables AI assistants to interact with Yamcs mission control systems through natural language, providing tools and resources for telemetry, commands, links, storage, instances, and alarms.
README
Yamcs MCP Server
A comprehensive Model Context Protocol (MCP) server for Yamcs (Yet Another Mission Control System) that exposes Yamcs capabilities as standardized MCP tools and resources.
Overview
The Yamcs MCP Server enables AI assistants to interact with mission control systems through natural language by providing a bridge between MCP-compatible clients and Yamcs instances. It implements the MCP protocol using FastMCP 2.x with a modular component architecture.
Features
- Mission Database (MDB): Access to parameters, commands, algorithms, and space systems
- TM/TC Processing: Real-time telemetry monitoring and command execution
- Link Management: Monitor and control data links
- Object Storage: Manage buckets and objects in Yamcs storage
- Instance Management: Control Yamcs instances and services
- Alarm Management: Monitor and acknowledge alarms with summary statistics
Installation
Prerequisites
- Python 3.12 or higher
- Yamcs server instance (local or remote)
- uv package manager (recommended) or pip
Using uv (recommended)
# Clone the repository
git clone https://github.com/PaulMRamirez/yamcs-mcp-server.git
cd yamcs-mcp-server
# Install dependencies
uv sync
# Run the server
uv run yamcs-mcp
Using pip
# Clone the repository
git clone https://github.com/PaulMRamirez/yamcs-mcp-server.git
cd yamcs-mcp-server
# Create virtual environment
python -m venv venv
source venv/bin/activate # On Windows: venv\Scripts\activate
# Install the package
pip install -e .
# Run the server
yamcs-mcp
Running Yamcs
You'll need a running Yamcs instance. The easiest way is using Docker:
docker run -d --name yamcs -p 8090:8090 yamcs/example-simulation
This will start Yamcs with a simulator instance that includes example telemetry data.
Configuration
The server can be configured using environment variables or a .env file:
# Yamcs connection settings
YAMCS_URL=http://localhost:8090
YAMCS_INSTANCE=simulator
YAMCS_USERNAME=admin
YAMCS_PASSWORD=password
# Server toggles
YAMCS_ENABLE_MDB=true
YAMCS_ENABLE_PROCESSOR=true
YAMCS_ENABLE_LINKS=true
YAMCS_ENABLE_STORAGE=true
YAMCS_ENABLE_INSTANCES=true
YAMCS_ENABLE_ALARMS=true
YAMCS_ENABLE_COMMANDS=true
# Server settings
MCP_TRANSPORT=stdio
MCP_HOST=127.0.0.1
MCP_PORT=8000
Usage
With Claude Desktop
Add the server to your Claude Desktop configuration:
{
"mcp-servers": {
"yamcs": {
"command": "uv",
"args": ["--directory", "/path/to/yamcs-mcp-server", "run", "yamcs-mcp"],
"env": {
"YAMCS_URL": "http://localhost:8090",
"YAMCS_INSTANCE": "simulator"
}
}
}
}
Important:
- Replace
/path/to/yamcs-mcp-serverwith the actual path to your yamcs-mcp-server directory - The
--directoryargument is required for uv to find the correct project - If
uvis not in your PATH, use the full path to uv (e.g.,/Users/PaulMRamirez/.local/bin/uv)
Available Tools
The server exposes numerous tools organized by server:
MDB Tools
mdb_list_parameters- List available parametersmdb_describe_parameter- Get parameter detailsmdb_list_commands- List available commandsmdb_describe_command- Get command detailsmdb_list_space_systems- List space systemsmdb_describe_space_system- Get space system details
Processor Tools
processors_list_processors- List available processorsprocessors_describe_processor- Get processor detailsprocessors_delete_processor- Delete a processorprocessors_issue_command- Issue a commandprocessors_subscribe_parameters- Subscribe to parameter updates
Link Tools
links_list_links- List all data linkslinks_describe_link- Get detailed link informationlinks_enable_link- Enable a data linklinks_disable_link- Disable a data link
Instance Tools
instances_list_instances- List Yamcs instancesinstances_describe_instance- Get instance detailsinstances_start_instance- Start an instanceinstances_stop_instance- Stop an instance
Storage Tools
storage_list_buckets- List storage bucketsstorage_list_objects- List objects in a bucketstorage_upload_object- Upload an objectstorage_download_object- Download an object
Alarm Tools
alarms_list_alarms- List active alarms with summary countsalarms_describe_alarm- Get detailed alarm informationalarms_acknowledge_alarm- Acknowledge an alarmalarms_shelve_alarm- Temporarily shelve an alarmalarms_unshelve_alarm- Unshelve an alarmalarms_clear_alarm- Clear an alarmalarms_read_log- Read alarm history
Command Tools
commands_list_commands- List available commands for executioncommands_describe_command- Get detailed command informationcommands_run_command- Execute a command (supports dry-run)commands_read_log- Read command execution history
Available Resources
The server also provides read-only resources:
mdb://parameters- List all parametersprocessors://list- List all processors with detailslinks://status- Show status of all linksinstances://list- List all instances with detailsalarms://list- Show active alarms summary
Development
Setting up the development environment
# Install development dependencies
uv sync --all-extras
# Install pre-commit hooks
pre-commit install
# Run tests
uv run pytest
# Run linting
uv run ruff check .
# Run type checking
uv run mypy src/
Testing
Run the test suite:
# Run all tests
uv run pytest
# Run with coverage
uv run pytest --cov=yamcs_mcp --cov-report=html
# Run specific test file
uv run pytest tests/test_server.py
# Run with verbose output
uv run pytest -v
Testing Without a Yamcs Server
The server can run in demo mode without connecting to a real Yamcs server:
# Run in demo mode (will show a warning about connection failure but continue)
uv run python -m yamcs_mcp.server
# Or use the demo script
uv run python run_demo.py
Project Structure
yamcs-mcp-server/
├── src/
│ └── yamcs_mcp/
│ ├── server.py # Main server entry point
│ ├── servers/ # MCP servers
│ │ ├── base_server.py # Base class for all servers
│ │ ├── mdb.py # Mission Database
│ │ ├── processors.py # TM/TC Processing
│ │ ├── links.py # Link management
│ │ ├── storage.py # Object storage
│ │ ├── instances.py # Instance management
│ │ └── alarms.py # Alarm management
│ ├── client.py # Yamcs client management
│ ├── config.py # Configuration
│ └── types.py # Type definitions
├── tests/ # Test suite
├── scripts/ # Test scripts
└── CLAUDE.md # AI assistant guidance
Troubleshooting
Common Issues
"Input validation error" when executing commands
Problem: Getting errors like '{"voltage_num": 1}' is not valid under any of the given schemas
Solution: The commands/run_command tool now accepts both formats. The server will automatically parse JSON strings to objects.
Both formats are now supported:
✅ Args as object (preferred):
{
"command": "/YSS/SIMULATOR/SWITCH_VOLTAGE_OFF",
"args": {"voltage_num": 1}
}
✅ Args as JSON string (automatically parsed):
{
"command": "/YSS/SIMULATOR/SWITCH_VOLTAGE_OFF",
"args": "{\"voltage_num\": 1}"
}
✅ Command without arguments:
{
"command": "/TSE/simulator/get_identification"
}
✅ Multiple arguments:
{
"command": "/YSS/SIMULATOR/SET_HEATER",
"args": {
"heater_id": 2,
"temperature": 25.5,
"duration": 300
}
}
Connection to Yamcs fails
Problem: Server can't connect to Yamcs at startup
Solution:
- Ensure Yamcs is running:
docker ps | grep yamcs - Check the URL is correct:
curl http://localhost:8090/api - The server will continue in demo mode even if Yamcs is unavailable
Enum serialization errors
Problem: Errors about unable to serialize enum types
Solution: This has been fixed in the latest version. Update to the latest version of the server.
Contributing
Contributions are welcome! Please read our Contributing Guidelines before submitting pull requests.
License
This project is licensed under the MIT License - see the LICENSE file for details.
Acknowledgments
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。