Prompt Store MCP
A local-only MCP server that uses SQLite to store, search, and manage a personal library of AI prompts. It enables developers to organize and reuse prompts across multiple AI clients like Claude and Cursor while keeping all data on their local machine.
README
@mhoshdev/prompt-store-mcp
A local-only MCP (Model Context Protocol) server for managing AI prompts with SQLite storage. Designed for individual developers to organize prompts without complex web services.
✨ Why Use This?
Your personal prompt library - Store, search, and reuse prompts across all your AI tools.
- Simple & focused - Just prompts, nothing more. No cloud, no accounts, no complexity.
- Works everywhere - Same prompts available in Claude, Cursor, Windsurf, any MCP client.
- 100% local - All data stays on your machine in SQLite. Your prompts, your control.
- Fast search - Find prompts by keyword or filter by tags instantly.
- Easy setup - One npx command, no install needed.
📦 Installation
No installation needed. Just use npx:
npx -y @mhoshdev/prompt-store-mcp@latest
Note: Requires Node.js 20+.
⚙️ MCP Client Configuration
Claude Desktop
Add to ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows):
{
"mcpServers": {
"prompt-store": {
"command": "npx",
"args": ["-y", "@mhoshdev/prompt-store-mcp@latest"]
}
}
}
Cursor IDE
Add to your Cursor MCP settings:
{
"mcpServers": {
"prompt-store": {
"command": "npx",
"args": ["-y", "@mhoshdev/prompt-store-mcp@latest"]
}
}
}
Windsurf
Add to your Windsurf MCP configuration:
{
"mcpServers": {
"prompt-store": {
"command": "npx",
"args": ["-y", "@mhoshdev/prompt-store-mcp@latest"]
}
}
}
For CLI access (e.g.,
--reset):npx -y @mhoshdev/prompt-store-mcp@latest --reset
CLI Options
# Start MCP server (default)
npx -y @mhoshdev/prompt-store-mcp@latest
# Reset database (clear all data, keep schema)
npx -y @mhoshdev/prompt-store-mcp@latest --reset
🛠️ Available Tools
| Tool | Description |
|---|---|
add_prompt |
Store a new prompt with optional tags |
list_prompts |
List prompts with pagination |
get_prompt |
Retrieve full prompt by ID |
update_prompt |
Update prompt title, content, and tags |
delete_prompt |
Permanently remove a prompt |
search_prompts |
Search by keyword in title/content |
filter_by_tags |
Filter prompts by tags (OR logic) |
list_tags |
List all tags with prompt counts |
💡 Example Usage
Store a Prompt
Use add_prompt with:
- title: "Code Review Assistant"
- content: "You are a senior code reviewer. Analyze code for bugs, security issues, and suggest improvements."
- tags: ["coding", "review"]
Search Prompts
Use search_prompts with:
- query: "review"
Filter by Tags
Use filter_by_tags with:
- tags: ["coding", "review"]
💾 Data Storage
- Location:
~/.prompt-store/prompts.db - Format: SQLite database
- Permissions: User-only read/write (600)
🔧 Development
# Install dependencies
pnpm install
# Build
pnpm build
# Run tests
pnpm test
# Type check
pnpm typecheck
🔍 Troubleshooting
Server won't start
- Ensure Node.js v20+ is installed:
node --version - Check database directory permissions:
ls -la ~/.prompt-store - View error logs: errors output to stderr
Database corruption
- Backup:
cp ~/.prompt-store/prompts.db ~/.prompt-store/prompts.db.backup - Reset:
npx -y @mhoshdev/prompt-store-mcp@latest --reset - Re-add prompts
MCP client can't connect
- Verify MCP client configuration path is correct
- Restart MCP client after configuration changes
- Ensure
npx -y @mhoshdev/prompt-store-mcp@latestruns successfully in terminal
License
MIT
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。