Fizzy

Provides programmable access to Fizzy data via MCP, enabling AI assistants to manage boards, cards, and projects.
  • javascript

6

GitHub Stars

javascript

Language

5 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

This MCP server lets AI assistants talk to Fizzy to manage boards, cards, and projects through natural language. It supports multiple transports, works locally or at the edge, and includes robust authentication, caching, and retries to help you automate workflows with your Fizzy data.

How to use

Choose an MCP client to connect to the Fizzy MCP server. You can run the server locally for personal use using stdio transport, or expose a remote HTTP endpoint for multi-user production deployments. Your client will send requests to the server to list boards, create cards, update columns, add comments, and more using the tools exposed by Fizzy.

How to install

Prerequisites: you need Node.js 18 or higher and a Fizzy account with API access.

Option A: Install from npm (recommended) and run locally with stdio transport.

npm install -g fizzy-mcp
"FIZZY_ACCESS_TOKEN"=your-token npx fizzy-mcp --transport stdio --port 3000

Additional configuration and usage notes

You can run the server in different modes depending on your deployment needs. For personal use and IDE integrations, stdio transport is ideal. For production deployments and multi-user scenarios, use HTTP transport with per-user tokens.

Configuration

Core environment variables you will encounter include FIZZY_ACCESS_TOKEN for stdio, and MCP_AUTH_TOKEN plus MCP_ALLOWED_ORIGINS and MCP_BIND_ALL_INTERFACES for HTTP/SSE transports. The server binds to localhost by default for safety, and per-user sessions time out after 30 minutes of inactivity.

Key configuration options shown in typical setups:

# Example: Local stdio setup
FIZZY_ACCESS_TOKEN=your-token npx fizzy-mcp --transport stdio --port 3000

# Example: Remote HTTP setup (production)
npx fizzy-mcp --transport http --port 3000
# Clients connect with per-user Authorization headers: Bearer <Fizzy token>

Security and authentication

Two authentication models are supported. For stdio (local) transports, a single user authenticates with FIZZY_ACCESS_TOKEN. For HTTP/SSE transports, each user provides their own Fizzy Personal Access Token via the Authorization header. Optionally, you can require client authentication with MCP_AUTH_TOKEN and restrict origins with MCP_ALLOWED_ORIGINS.

Note: When deploying to Cloudflare Workers, only HTTP transport is supported and tokens are per-user.

Troubleshooting

If the server cannot be discovered by your IDE, ensure you have restarted after configuration changes and that Node.js is installed. For token or 404 errors, verify your Fizzy token is valid and that you have access to the correct account.

If you see connection issues, you can test your token directly against Fizzy before connecting through MCP.

Notes and tips

Live server to try quickly: use the live endpoint at https://fizzy.fabric.pro/mcp to connect with your MCP client and tokens.

Avoid SSE for new deployments as it is deprecated; prefer HTTP transport for new projects.

Troubleshooting prompts and common scenarios

If you want to test with an MCP Inspector, run the HTTP server and connect Inspector with the HTTP transport, pointing to http://localhost:3000/mcp and using your Fizzy token in the Authorization header.

Available tools

fizzy_get_identity

Get current user identity and accounts

fizzy_get_accounts

List all accessible Fizzy accounts

fizzy_get_account

Get details of a specific account

fizzy_get_boards

List boards in an account

fizzy_get_board

Get details of a specific board

fizzy_create_board

Create a new board

fizzy_update_board

Update a board's name

fizzy_delete_board

Delete a board

fizzy_get_cards

List cards with optional filters

fizzy_get_card

Get card details

fizzy_create_card

Create a new card with title, description, status, column, assignees, tags, due date

fizzy_update_card

Update any card property

fizzy_delete_card

Delete a card

fizzy_close_card

Close a card (mark as done)

fizzy_reopen_card

Reopen a closed card

fizzy_move_card_to_not_now

Move a card to Not Now triage

fizzy_move_card_to_column

Move a card to a specific column

fizzy_send_card_to_triage

Send a card back to triage (remove from column)

fizzy_toggle_card_tag

Toggle a tag on/off for a card

fizzy_toggle_card_assignment

Toggle a user assignment on/off for a card

fizzy_watch_card

Subscribe to notifications for a card

fizzy_unwatch_card

Unsubscribe from notifications for a card

fizzy_get_card_comments

List comments on a card

fizzy_get_comment

Get a specific comment

fizzy_create_comment

Add a comment to a card (supports HTML)

fizzy_update_comment

Update a comment

fizzy_delete_comment

Delete a comment

fizzy_get_reactions

Get all emoji reactions on a comment

fizzy_add_reaction

Add an emoji reaction to a comment

fizzy_remove_reaction

Remove an emoji reaction from a comment

fizzy_get_step

Get a specific to-do step on a card

fizzy_create_step

Create a new to-do step on a card

fizzy_update_step

Update a step (description or completion status)

fizzy_delete_step

Delete a step from a card

fizzy_get_columns

List columns on a board

fizzy_get_column

Get column details

fizzy_create_column

Create a new column with name and color

fizzy_update_column

Update column name/color

fizzy_delete_column

Delete a column

fizzy_get_tags

List all tags in an account

fizzy_get_users

List users in an account

fizzy_get_user

Get user details

fizzy_update_user

Update user's display name

fizzy_deactivate_user

Deactivate a user

fizzy_get_notifications

List notifications for current user

fizzy_mark_notification_read

Mark notification as read

fizzy_mark_notification_unread

Mark notification as unread

fizzy_mark_all_notifications_read

Mark all notifications as read

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational