mcpField
K
karea-mcp

karea-mcp

Local
karea.appProductivity1Updated 6d ago

About karea-mcp

karea-mcp connects MCP-compatible clients to Karea for managing tasks, projects, notes, meetings, and related work. Its nine tools cover 70 actions, including task links to Jira and productivity recaps.

Key features

  1. 01Create, edit, close, and delete tasks
  2. 02Manage task notes and markdown
  3. 03Track questions, meetings, and reminders
  4. 04Attach resources and link Jira tasks
  5. 05View productivity recaps

Use cases

  1. 01Update tasks while coding
  2. 02Keep design notes with a task
  3. 03Track open questions and meeting agendas
  4. 04Review completed work over a date range

MCP support

Exposes Karea task-management capabilities to MCP-compatible clients through nine tools covering 70 actions.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

mcp.json · Universal config
{
"mcpServers": {
"starecz-karea": {
"command": "npx",
"args": [
"-y",
"karea-mcp"
],
"env": {
"KAREA_API_KEY": "<KAREA_API_KEY>",
"KAREA_URL": "<KAREA_URL>"
}
}
}
}
Run commandnpx -y karea-mcp
Environment variables to set
KAREA_API_KEYSecretYour Karea API key. Generate at https://karea.app/dashboard/settings?section=api-keys
KAREA_URLKarea base URL. Defaults to https://karea.app for the hosted service.

Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.

Packages
npmkarea-mcp0.14.0

Tools

9

Saved tool list · updated 5d ago

ParameterTypeRequiredDescription
actionenumNoWhich operation to perform: one of the actions listed in this tool's description (an action name, not a tool name).karea_list_projects | karea_list_tasks | karea_view_task | karea_view_tasks | karea_task_changes | karea_list_subtasks | karea_list_notes | karea_list_sticky_notes | karea_get_markdown | karea_get_context | karea_list_questions | karea_list_resources | karea_get_resource | karea_list_meetings | karea_view_meeting | karea_check_reminders | karea_get_jira_link | karea_list_sessions | karea_recap
paramsobjectNoArguments for the chosen action, as an object. Omit for actions that take none.

Details

From the project README

karea-mcp

The MCP server for Karea. 9 tools so Claude Code, Cursor, and any other MCP-compatible client can create, edit, close, recap, link, and otherwise manage your tasks while you code.

npm license karea-mcp MCP server

What you can do with it

  • Create / edit / close / delete tasks and subtasks from inside Claude Code while it codes
  • Read and write long-form markdown on any task (perfect for solution design + investigation notes)
  • Track open questions per task and answer them later
  • Attach resources (file references, links) to tasks
  • Link tasks to Jira and back
  • Get a productivity recap across a date range
  • Manage categories, projects, and project sharing

Install

npx karea-mcp

Configure (Claude Code)

Add to ~/.claude.json (or run claude mcp add):

{
  "mcpServers": {
    "karea": {
      "command": "npx",
      "args": ["-y", "karea-mcp"],
      "env": {
        "KAREA_API_KEY": "your_key_here",
        "KAREA_URL": "https://karea.app"
      }
    }
  }
}

Get your API key at https://karea.app/dashboard/settings?section=api-keys.

Configure (Cursor)

Add to .cursor/mcp.json in your project, or to ~/.cursor/mcp.json globally:

{
  "mcpServers": {
    "karea": {
      "command": "npx",
      "args": ["-y", "karea-mcp"],
      "env": {
        "KAREA_API_KEY": "your_key_here"
      }
    }
  }
}

Tool catalogue

9 tools covering 70 actions (regenerated by scripts/sync-metadata.mjs).

Each tool takes an action and a params object. Actions are not tools: karea_list_notes is called as karea_read with { "action": "karea_list_notes", "params": { ... } } (the karea_ prefix of an action is optional). Set KAREA_MCP_LEGACY_TOOLS=1 to go back to registering all 70 actions as individual tools.

  • karea_read - Read anything in Karea without changing it: projects, tasks, subtasks, notes, sticky notes, task documents and AI context, open questions, resources, meetings, reminders, Jira links, AI sessions, and the activity recap. The main entry point - start here.
    • karea_list_projects, karea_list_tasks, karea_view_task, karea_view_tasks, karea_task_changes, karea_list_subtasks, karea_list_notes, karea_list_sticky_notes, karea_get_markdown, karea_get_context, karea_list_questions, karea_list_resources, karea_get_resource, karea_list_meetings, karea_view_meeting, karea_check_reminders, karea_get_jira_link, karea_list_sessions, karea_recap
  • karea_tasks_write - Create and change tasks: create, edit one or many, close one or many, log finished or in-progress work, plus subtasks and closing requisites (the checklist a task needs before it can close).
    • karea_create_task, karea_quick_task, karea_doing, karea_edit_task, karea_edit_tasks, karea_close_task, karea_done, karea_create_subtask, karea_add_requisite, karea_toggle_requisite
  • karea_notes_write - Write notes: add or edit task notes (human-readable updates) and sticky notes, and write a task's markdown document or an entry of its AI Context (private cross-session memory). Overwrites what an edit replaces.
    • karea_add_note, karea_edit_note, karea_create_sticky_note, karea_edit_sticky_note, karea_set_markdown, karea_set_context
  • karea_projects_write - Create a project or a category inside one, or share a project with someone by email.
    • karea_create_project, karea_create_category, karea_share_project
  • karea_meetings_write - Create and edit meetings and put tasks or questions on their agenda; create, answer and edit open questions; create, snooze, dismiss or complete reminders.
    • karea_create_meeting, karea_edit_meeting, karea_link_task_to_meeting, karea_link_question_to_meeting, karea_create_question, karea_answer_question, karea_edit_question, karea_create_reminder, karea_snooze_reminder, karea_dismiss_reminder, karea_mark_reminder_done
  • karea_resources_write - Create, upload and update resources in the file/document library and attach them to tasks; link a task to a Jira issue or to an AI coding session.
    • karea_create_resource, karea_upload_resource, karea_update_resource, karea_link_resource_to_task, karea_link_jira, karea_link_session
  • karea_assistant - Send a natural-language request to the Karea AI assistant, which may read or change your tasks to carry it out.
    • karea_ask
  • karea_delete - Delete or detach things. Permanent deletes (project, category, task, requisite, note, sticky note, question, resource, meeting) need confirm=true where stated; detaching a resource, task, question, Jira issue or AI session from what it is linked to.
    • karea_delete_project, karea_delete_category, karea_delete_task, karea_delete_requisite, karea_delete_note, karea_delete_sticky_note, karea_delete_question, karea_delete_resource, karea_delete_meeting, karea_unlink_resource_from_task, karea_unlink_task_from_meeting, karea_unlink_question_from_meeting, karea_unlink_jira, karea_unlink_session
  • karea_help - full parameter schema for any action.

List actions need at least one filter, so a single call cannot dump everything you have:

ActionAccepted filters
karea_list_tasksprojectId, status, closedSince, category, priority, assignee, search
karea_list_questionsprojectId, status
karea_list_meetingsscope, from, to, projectId
karea_list_resourcesprojectId, query, folder, type, mime, minSize, maxSize
karea_list_sticky_notesprojectId, or scope (global / all)

A call without one is refused with the accepted filters listed, so the agent can retry at once. karea_list_projects needs none (it is how an agent finds the projectId), and neither do pending reminders in karea_check_reminders.

Workflow example

You: I'm starting on the auth bug.
Claude (via karea-mcp): karea_doing("auth bug")
        --> status set to in_progress
You: Add a closing requisite: must pass the failing test.
Claude: karea_add_requisite(...)
You (later): I fixed it.
Claude: karea_toggle_requisite(...) then karea_done("auth bug")

License

MIT. See LICENSE.

Comments

Markdown supported

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.