Statespace
Enables users to document data and connect it to AI agents by defining tools and instructions in markdown files. It supports building RAG and text-to-SQL applications that can be deployed as MCP servers, APIs, or CLIs.
README
<br>
<div align="center"> <picture> <source media="(prefers-color-scheme: light)" srcset="https://raw.githubusercontent.com/statespace-tech/statespace/main/docs/assets/images/header_light.png" /> <img src="https://raw.githubusercontent.com/statespace-tech/statespace/main/docs/assets/images/header_dark.png" alt="Statespace" width="375" /> </picture> </div>
<div align="center">
<br>
Self-documenting AI applications
</div>
Website: https://statespace.com
Documentation: https://docs.statespace.com
AI doesn't know your data. Statespace helps you build self-documenting data applications that describe themselves to agents. Build RAG, text-to-SQL, and knowledge bases that agents can maintain and improve over time. Once you’ve created an app, you can deploy, manage, and share it from our cloud platform.
Quickstart
Get up and running with your first app by asking your coding agent:
claude "Create a Statespace app: https://statespace.com/AGENTS.md"
Example
1. Create it
Create a new PostgreSQL app with:
statespace init --text-to-sql postgresql://user:pass@host:port/db
This creates an app skeleton with just enough tools and instructions for your agent to explore your database:
---
tools:
- [psql, -d, $DB, -c, { regex: "^SELECT\\b.*" }]
---
# Instructions
- Learn the schema by exploring tables, columns, and relationships
- Translate the user's question into a query that answers it
2. Build it
Build your full app by iterating on it with your coding agent:
claude "Document my database's schema and add a script to summarize them"
Your agent will spin up a local server to test your app:
statespace serve .demo/ --port 8000
The result is a self-documenting app that any agent can understand and use:
demo/
├── README.md # from above
├── script.py
└── schema/
├── users.md
└── products.md
3. Ship it
Optionally, create a free Statespace account to deploy your app and get a shareable URL:
statespace deploy path/to/app/
Then, connect any agent to your app through our API or MCP.
<details open> <summary>API</summary>
Pass the public URL to your agent:
claude "Use the API at https://demo.statespace.app to find out the number of users"
</details>
<details> <summary>MCP</summary>
Configure the MCP client:
"statespace": {
"command": "uvx",
"args": ["statespace-mcp", "https://demo.statespace.app"]
}
</details>
Example skeletons
Community & Contributing
- Discord: Join our community server for real-time help and discussions
- X: Follow us @statespace_tech for updates and news
- Issues: Report bugs or request features on GitHub Issues
License
This project is licensed under the terms of the MIT license.
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。