MCP Toolkit
A collection of production-ready MCP servers for PostgreSQL, SQLite, Redis, File System, and GitHub API, enabling database operations, file management, and GitHub interactions through natural language.
README
MCP Toolkit
A comprehensive collection of production-ready Model Context Protocol (MCP) servers for common tools and services.
Features
- PostgreSQL Server - Query, insert, update, and manage PostgreSQL databases
- SQLite Server - Lightweight SQLite database operations with full SQL support
- Redis Server - Key-value operations, pub/sub, and data structure manipulation
- File System Server - Secure file operations with sandboxing and access controls
- GitHub API Server - Repository management, issues, PRs, and more via GitHub API
Installation
# Install from PyPI
pip install mcp-toolkit
# Or install specific servers only
pip install mcp-toolkit[postgres]
pip install mcp-toolkit[sqlite]
pip install mcp-toolkit[redis]
pip install mcp-toolkit[filesystem]
pip install mcp-toolkit[github]
# Install all servers
pip install mcp-toolkit[all]
Quick Start
PostgreSQL
from mcp_toolkit.servers.postgres import PostgreSQLServer
server = PostgreSQLServer(
connection_string="postgresql://user:pass@localhost:5432/mydb"
)
server.run()
SQLite
from mcp_toolkit.servers.sqlite import SQLiteServer
server = SQLiteServer(database_path="./data.db")
server.run()
Redis
from mcp_toolkit.servers.redis import RedisServer
server = RedisServer(redis_url="redis://localhost:6379/0")
server.run()
File System
from mcp_toolkit.servers.filesystem import FilesystemServer
server = FilesystemServer(
root_dir="/path/to/workspace",
read_only=False
)
server.run()
GitHub API
from mcp_toolkit.servers.github import GitHubServer
server = GitHubServer(
github_token="ghp_..."
)
server.run()
CLI Usage
# Run PostgreSQL server
mcp-toolkit postgres --connection-string "postgresql://localhost/mydb"
# Run SQLite server
mcp-toolkit sqlite --database ./data.db
# Run Redis server
mcp-toolkit redis --url redis://localhost:6379
# Run Filesystem server
mcp-toolkit filesystem --root ./workspace
# Run GitHub server
mcp-toolkit github --token $GITHUB_TOKEN
Configuration
Environment Variables
# PostgreSQL
POSTGRES_CONNECTION_STRING=postgresql://user:pass@localhost/db
# Redis
REDIS_URL=redis://localhost:6379/0
# GitHub
GITHUB_TOKEN=ghp_xxxxxxxxxxxx
Claude Desktop Configuration
Add to your Claude Desktop config (claude_desktop_config.json):
{
"mcpServers": {
"postgres": {
"command": "mcp-toolkit",
"args": ["postgres", "--connection-string", "postgresql://localhost/mydb"]
},
"redis": {
"command": "mcp-toolkit",
"args": ["redis", "--url", "redis://localhost:6379"]
}
}
}
API Reference
PostgreSQL Server
| Tool | Description |
|---|---|
query |
Execute a SELECT query |
execute |
Execute an INSERT/UPDATE/DELETE statement |
list_tables |
List all tables in the database |
describe_table |
Get table schema information |
explain_query |
Get query execution plan |
SQLite Server
| Tool | Description |
|---|---|
query |
Execute a SELECT query |
execute |
Execute an INSERT/UPDATE/DELETE statement |
list_tables |
List all tables |
describe_table |
Get table schema |
vacuum |
Optimize database storage |
Redis Server
| Tool | Description |
|---|---|
get |
Get a key's value |
set |
Set a key-value pair |
delete |
Delete one or more keys |
keys |
List keys matching a pattern |
hget |
Get a hash field value |
hset |
Set a hash field |
lrange |
Get a list range |
publish |
Publish a message to a channel |
Filesystem Server
| Tool | Description |
|---|---|
read_file |
Read file contents |
write_file |
Write content to a file |
list_directory |
List directory contents |
create_directory |
Create a new directory |
delete_file |
Delete a file |
move_file |
Move/rename a file |
search_files |
Search for files by pattern |
GitHub Server
| Tool | Description |
|---|---|
list_repos |
List user repositories |
get_repo |
Get repository details |
list_issues |
List repository issues |
create_issue |
Create a new issue |
list_prs |
List pull requests |
get_file |
Get file contents from repo |
search_code |
Search code across repos |
Development
# Clone the repository
git clone https://github.com/exycfs/mcp-toolkit.git
cd mcp-toolkit
# Install development dependencies
pip install -e ".[dev]"
# Run tests
pytest
# Run linters
ruff check .
mypy src/
# Format code
ruff format .
Contributing
Contributions are welcome! Please read our Contributing Guidelines before submitting a Pull Request.
License
This project is licensed under the MIT License - see the LICENSE file for details.
Acknowledgments
- Built on the Model Context Protocol specification
- Inspired by the MCP ecosystem and community
Support
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。