About GanttFather
GanttFather MCP server connects AI assistants to GanttFather, an online Gantt chart and Kanban tool, to read and edit project tasks, dates, dependencies, resources, and critical paths. The hosted service requires no local installation.
Key features
- 01Read and edit project tasks
- 02Manage dates and task dependencies
- 03Inspect critical paths and project progress
- 04Manage project resources
- 05Render Gantt charts or interactive views
Use cases
- 01Review project tasks and timelines
- 02Update task dates or assignees
- 03Check task dependencies and critical paths
MCP support
The remote MCP server exposes tools for reading and editing GanttFather projects over Streamable HTTP. Authentication uses OAuth 2.1 or a project token.
Configuration
Remote servers need no local install. Pick your client, then copy the config or install in one click.
https://mcp.ganttfather.com/claude mcp add --transport http com-ganttfather-ganttfather https://mcp.ganttfather.com/{"mcpServers":{"com-ganttfather-ganttfather":{"type":"http","url":"https://mcp.ganttfather.com/"}}}Open in Cursor{"servers":{"com-ganttfather-ganttfather":{"type":"http","url":"https://mcp.ganttfather.com/"}}}Open in VS Code{"mcpServers":{"com-ganttfather-ganttfather":{"type":"http","url":"https://mcp.ganttfather.com/"}}}{"mcpServers":{"com-ganttfather-ganttfather":{"serverUrl":"https://mcp.ganttfather.com/"}}}Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.
Tools
Details
From the project READMEGanttFather MCP server
A remote Model Context Protocol server for GanttFather, an online Gantt chart and Kanban tool. It lets an AI assistant read and edit your GanttFather projects: tasks, dates, dependencies, resources, the critical path, and the chart itself.
This repository holds documentation and the registry entry only. The server is hosted; there is nothing to install or run.
- URL:
https://mcp.ganttfather.com/ - Transport: streamable HTTP
- Sign-in: OAuth 2.1 with dynamic client registration and PKCE (S256), scopes
mcp:readandmcp:write. Clients that can't do OAuth can send a project token from GanttFather (AI Agents & API) asAuthorization: Bearer <token>. - Official MCP Registry:
com.ganttfather/ganttfather - Docs and setup guides: https://ganttfather.com/mcp/
- Price: free for one project with two editors; viewers and guests are free.
Add it to your client
The quickest way is to add the URL as a remote server and sign in when your client asks.
Claude (claude.ai and the desktop app): Settings, Connectors, Add custom connector, then paste https://mcp.ganttfather.com/.
Claude Code
claude mcp add --transport http ganttfather https://mcp.ganttfather.com/
VS Code (GitHub Copilot Chat), in .vscode/mcp.json:
{ "servers": { "ganttfather": { "type": "http", "url": "https://mcp.ganttfather.com/" } } }
GitHub Copilot CLI
copilot mcp add --transport http ganttfather https://mcp.ganttfather.com/
Then run /mcp auth ganttfather inside Copilot CLI.
Cursor, in ~/.cursor/mcp.json:
{ "mcpServers": { "ganttfather": { "url": "https://mcp.ganttfather.com/" } } }
Windsurf and Antigravity, in mcp_config.json (they read serverUrl):
{ "mcpServers": { "ganttfather": { "serverUrl": "https://mcp.ganttfather.com/" } } }
Codex
codex mcp add ganttfather --url https://mcp.ganttfather.com/
Gemini CLI, in settings.json (it needs httpUrl; a plain url means an older transport there):
{ "mcpServers": { "ganttfather": { "httpUrl": "https://mcp.ganttfather.com/" } } }
OpenCode, in opencode.json, then run opencode mcp auth ganttfather:
{ "$schema": "https://opencode.ai/config.json", "mcp": { "ganttfather": { "type": "remote", "url": "https://mcp.ganttfather.com/", "enabled": true } } }
Guides for ChatGPT, Gemini, Goose, n8n and others are at https://ganttfather.com/mcp/. Of the clients above, we have tested Claude end to end; reports from the others are welcome in Issues.
Tools
27 tools. Every tool sets readOnlyHint, destructiveHint and openWorldHint explicitly; writes that can overwrite or remove data are marked destructive, so your client can ask before running them.
| Tool | Title | Kind |
|---|---|---|
draft_support_email | Draft support email (never sends it) | read |
get_critical_path | Get critical path | read |
get_overdue_tasks | Get overdue tasks | read |
get_project | Get project | read |
get_project_summary | Get project summary | read |
get_project_timeline | Get project timeline | read |
get_task | Get task | read |
list_dependencies | List dependencies | read |
list_members | List members | read |
list_projects | List projects | read |
list_resources | List resources | read |
list_tasks | List tasks | read |
render_gantt_image | Render Gantt image (SVG or Mermaid) | read |
show_gantt_chart | Show Gantt chart (interactive widget) | read |
add_dependency | Add dependency | write, destructive |
add_resource | Add resource | write |
add_task_image | Add image to task | write |
create_task | Create task | write |
delete_resource | Delete resource | write, destructive |
delete_task | Delete task | write, destructive |
remove_dependency | Remove dependency | write, destructive |
reorder_tasks | Reorder tasks | write, destructive |
set_task_assignees | Set task assignees | write, destructive |
set_task_dependencies | Set task dependencies | write, destructive |
update_project_context | Update project context | write, destructive |
update_resource | Update resource | write, destructive |
update_task | Update task | write, destructive |
Privacy and support
The assistant sees only the projects the signed-in account can see, and you can revoke its access at any time in GanttFather, under AI Agents & API. Privacy policy: https://ganttfather.com/privacy/. Support: [email protected].
License
The documentation and configuration in this repository are under the MIT License. The GanttFather service itself is proprietary.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.