Traggo

The server is stateless — it stores nothing. The Traggo token is the entire auth state.

┌─────────────┐ Bearer <token> ┌──────────────┐ traggo <token> ┌────────────┐ MCP client │ POST /mcp │ ─────────────────► │ traggo-mcp │ ─────────────────► │ Traggo │ └─────────────┘ │ (this server)│ GraphQL │ instance │ └──────────────┘ └────────────┘ browser ── GET/POST /login ──────────────► returns a device token to copy

Tags are key/value pairs (e.g.{"key": "project", "value": "traggo"}). Times are RFC3339 (e.g.2026-06-28T09:00:00Z).

The server needs one environment variable:TRAGGO_URL, the base URL of your Traggo instance (no trailing slash).

export TRAGGO_URL=https://traggo.example.com go run ./cmd/server

Use a streamable-HTTP MCP transport pointed at/mcp, with the token as a bearer credential. For example, in a client that supports HTTP MCP servers:

{ "mcpServers": { "traggo": { "url": "http://localhost:8080/mcp", "headers": { "Authorization": "Bearer <your-traggo-token>" } } } }

export TRAGGO_URL=https://traggo.example.com air

templ generate # .templ → _templ.go bunx @tailwindcss/cli -i ./app.css -o assets/output.css # build CSS go run ./cmd/server

cmd/server/ Entry point: route wiring, starts the HTTP server internal/ auth/ Web login flow → Traggo device token mcp/ MCP server: tool registration + bearer-token middleware traggo/ GraphQL client for the Traggo API (all timer/timespan/tag ops) views/ templ components and pages for the web UI assets/ Static files (generated CSS, images) .dagger/ Dagger module for building & publishing the container image

dagger call build \ --source . \ --registry registry.example.com \ --username you@example.com \ --registry-password env:REGISTRY_PASSWORD \ --image traggo-mcp:latest

--sourcedefaults to.and--imagedefaults totraggo-mcp:latest. The function returns the published image reference.

Contributions are welcome. To get started:
- Fork and clone the repo.
- Install the prerequisites above and runair(or the manual build steps) against a Traggo instance you control.
- Make your change. If you touch a.templfile, runtempl generate. If you add or change a dependency, rungo mod tidy.
- Make surego build ./...andgo vet ./...pass.
- Open a pull request describing the change.

- Define the input/output types ininternal/traggo/types.go(usejsonschemastruct tags so the MCP schema is descriptive).
- Add the operation method tointernal/traggo/service.go(it goes through the sharedexecutehelper for auth and error handling).
- Register the tool ininternal/mcp/server.go.

The Traggo GraphQL schema is the source of truth for available operations and field names.

Interact with task, doc, and project data in Dart, an AI-native project management tool

Remote MCP server for MeisterTask. Create and manage projects, tasks, and notes from your AI assistant. Hosted (streamable-HTTP) — connect at https://mcp.meistertask.com/mcp

Remote MCP server for MindMeister. Create, edit, and organize mind maps from your AI assistant. Hosted (streamable-HTTP) — connect at https://mcp.mindmeister.com/mcp

The official Plane MCP server provides integration with Plane APIs, enabling full AI automation of Plane projects, work items, cycles and more.

MCP server to interact with Routine: calendars, tasks, notes, etc.

Keep teams & agents coordinated automatically

From the creators of Wunderlist — the all-in-one task management app for to-do lists, notes, and projects. AI-powered productivity that replaces 5 apps.

Connect to the Taskade platform via MCP. Access tasks, projects, workflows, and AI agents in real-time through a unified workspace and API.