mcp-baserow-schema
Enables automated schema management for Baserow, including creating, updating, and deleting tables and fields, with built-in 2FA TOTP authentication.
README
mcp-baserow-schema
MCP server for Baserow that provides schema management with 2FA authentication (TOTP).
Built for workflows where schema changes (adding/removing/modifying fields and tables) must be automated — not manual. Pairs with Baserow's official MCP (data-only) for full coverage.
Why
Baserow's official MCP handles data CRUD but not schema changes. In 2026, plain password auth without2FA is not acceptable. This MCP solves both:
- Schema operations: create/update/delete fields and tables via MCP tools
- 2FA support: automatic TOTP-based authentication — no manual token management
- Designed for agents: AI agents can modify table structure without human intervention
Authentication
Supports Baserow's two-step2FA flow:
POST /api/user/token-auth/→ temporary2FA tokenPOST /api/two-factor-auth/verify/→ JWT access token (with TOTP code)
JWT is auto-refreshed. Credentials are passed via environment variables — never hardcoded.
Tools (16)
| Category | Tool | Description |
|---|---|---|
| Discovery | list_databases |
List databases in a workspace |
list_tables |
List tables in a database | |
| Schema | create_table |
Create a new table |
delete_table |
Delete a table | |
list_fields |
List fields (columns) in a table | |
create_field |
Create a new field | |
update_field |
Update an existing field | |
delete_field |
Delete a field | |
| Data | list_rows |
List rows with pagination/search/sort |
create_row |
Create a single row | |
update_row |
Update a single row | |
delete_row |
Delete a single row | |
batch_create_rows |
Create multiple rows (up to 200) | |
batch_update_rows |
Update multiple rows (up to 200) | |
batch_delete_rows |
Delete multiple rows (up to 200) | |
| Auth | auth_status |
Check authentication state |
Setup
Prerequisites
- Node.js ≥ 20
- Baserow account with2FA enabled
- Baserow TOTP secret (base32)
Install
git clone git@github.com:aficiomaquinas/mcp-baserow-schema.git
cd mcp-baserow-schema
npm install
npm run build
Configure
Set environment variables (or use a .env file):
BASEROW_API_URL=https://your-baserow-instance.com
BASEROW_USERNAME=you@example.com
BASEROW_PASSWORD=your_password
BASEROW_TOTP_SECRET=YOUR_BASE32_TOTP_SECRET
Hermes Agent
Add to ~/.hermes/profiles/<profile>/config.yaml:
mcp_servers:
baserow-mcp:
command: node
args:
- /path/to/mcp-baserow-schema/dist/index.js
enabled: true
env:
BASEROW_API_URL: https://baserow.ttamayo.com
BASEROW_USERNAME: you@example.com
BASEROW_PASSWORD: your_password
BASEROW_TOTP_SECRET: YOUR_BASE32_TOTP_SECRET
Claude Desktop
Add to claude_desktop_config.json:
{
"mcpServers": {
"baserow-schema": {
"command": "node",
"args": ["/path/to/mcp-baserow-schema/dist/index.js"],
"env": {
"BASEROW_API_URL": "https://baserow.ttamayo.com",
"BASEROW_USERNAME": "you@example.com",
"BASEROW_PASSWORD": "your_password",
"BASEROW_TOTP_SECRET": "YOUR_BASE32_TOTP_SECRET"
}
}
}
}
Field Types
The create_field and update_field tools support all Baserow field types:
text, long_text, url, email, number, rating, boolean, date, last_modified, last_modified_by, created_on, created_by, duration, link_row, file, single_select, multiple_select, phone_number, formula, count, rollup, lookup, multiple_collaborators, uuid, autonumber, password, ai
Usage with Official Baserow MCP
Use both MCPs together:
- Official Baserow MCP → data operations (optimized for row CRUD)
- mcp-baserow-schema → schema operations (fields, tables, structure)
This separation keeps each tool focused and avoids conflicts.
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。