fizzy-mcp
MCP server for Fizzy, Basecamp's open-source Kanban tool. Enables AI assistants to manage boards, cards, tags, columns, and comments via API.
README
Fizzy-MCP
An MCP (Model Context Protocol) server for Fizzy, Basecamp's open-source Kanban tool.
What is MCP?
Model Context Protocol lets AI assistants (like Claude) use external tools. This server gives Claude the ability to manage your Fizzy boards and cards.
Quick Start
1. Install
git clone https://github.com/clawdcraft/fizzy-mcp.git
cd fizzy-mcp
npm install
npm run build
2. Get a Fizzy API Token
- Open your Fizzy instance (e.g.,
http://localhost:3000) - Click your avatar → My profile
- Go to Access Tokens
- Click Generate a new access token
- Enter a description (e.g., "MCP Server")
- Select Read + Write permission
- Copy the token (you won't see it again!)
3. Configure Your Claude Client
Claude Code (macOS/Linux)
Edit ~/.claude.json:
{
"mcpServers": {
"fizzy": {
"command": "node",
"args": ["/full/path/to/fizzy-mcp/dist/index.js"],
"env": {
"FIZZY_URL": "http://localhost:3000",
"FIZZY_TOKEN": "your-api-token-here",
"FIZZY_ACCOUNT_ID": "1"
}
}
}
}
Claude Desktop (macOS)
Edit ~/Library/Application Support/Claude/claude_desktop_config.json:
{
"mcpServers": {
"fizzy": {
"command": "node",
"args": ["/full/path/to/fizzy-mcp/dist/index.js"],
"env": {
"FIZZY_URL": "http://localhost:3000",
"FIZZY_TOKEN": "your-api-token-here",
"FIZZY_ACCOUNT_ID": "1"
}
}
}
}
Then restart Claude.
4. Use It
Ask Claude things like:
- "List my Fizzy boards"
- "Create a card called 'Fix login bug' on the Ravenus board"
- "Move card 15 to Done"
- "Add a comment to card 12"
Environment Variables
| Variable | Description | Default |
|---|---|---|
FIZZY_URL |
Your Fizzy instance URL | http://localhost:3000 |
FIZZY_TOKEN |
API access token | (required) |
FIZZY_ACCOUNT_ID |
Account ID (from URL path) | 1 |
Available Tools
Boards
| Tool | Description |
|---|---|
fizzy_list_boards |
List all boards in the account |
fizzy_get_board |
Get details of a specific board |
fizzy_create_board |
Create a new board |
Cards
| Tool | Description |
|---|---|
fizzy_list_cards |
List all cards (optionally filter by board_id) |
fizzy_get_card |
Get card details |
fizzy_create_card |
Create a new card |
fizzy_update_card |
Update card title/description |
fizzy_move_card |
Move to column, "done", or "not_now" |
Tags
| Tool | Description |
|---|---|
fizzy_add_tag |
Add a tag to a card (creates tag if needed) |
fizzy_remove_tag |
Remove a tag from a card |
Columns & Comments
| Tool | Description |
|---|---|
fizzy_list_columns |
List columns on a board |
fizzy_add_comment |
Add a comment to a card |
fizzy_list_comments |
List comments on a card |
Examples
Create a card with HTML description
Create a card on board "03fjavj4afm5qrvuvluoi0po9" with:
- Title: "🔧 Fix database connection"
- Description with a link to the GitHub issue
The MCP server supports HTML in descriptions:
Fix the connection pooling issue.
📄 <a href="https://github.com/example/repo/issues/42">Issue #42</a>
Move cards through workflow
Move card 15 to "done"
Move card 12 to "not_now"
Tag cards
Add tag "bug" to card 15
Remove tag "feature" from card 12
Tags are created automatically if they don't exist.
Development
# Install dependencies
npm install
# Build
npm run build
# Watch mode
npm run dev
Testing Manually
export FIZZY_URL="http://localhost:3000"
export FIZZY_TOKEN="your-token"
export FIZZY_ACCOUNT_ID="1"
# List tools
echo '{"jsonrpc":"2.0","id":1,"method":"tools/list"}' | node dist/index.js
# Call a tool
echo '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"fizzy_list_boards","arguments":{}}}' | node dist/index.js
Troubleshooting
406 Not Acceptable
- Check that
FIZZY_TOKENis set and valid - Verify the token has Read + Write permissions
Connection refused
- Ensure Fizzy is running at the configured URL
- Check the port matches your Fizzy setup
Card not found
- Card IDs are the card number (e.g.,
11), not the CUID
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。