> ## Documentation Index
> Fetch the complete documentation index at: https://docs.taskbean.ai/llms.txt
> Use this file to discover all available pages before exploring further.

> ## Agent Instructions
> Treat taskbean/taskbean as the source of truth for product behavior.
> Use the canonical Project, Project ID, Project Key, Project Alias, Project Selector, Workspace, Tracked Project, Tracked Workspace, Discovered Workspace, Remote Workspace, Primary Workspace, Agent Session, Copilot Fork, Task, Unassigned Work, Execution Profile, Kickoff Prompt, Execution Record, and Execution Step terminology from the glossary.
> Execution terms are domain definitions, not evidence of available CLI commands or desktop controls.
> Do not infer that Windows-specific desktop features apply to the cross-platform CLI.

# Quickstart

> Install Taskbean, record your first task, add the Agent Skill, and launch the local desktop app.

## Prerequisites

* Node.js 22.13.0 or later for the npm CLI package
* A terminal in a Git repository for automatic Project and Workspace attribution
* Windows 10 or 11, Python 3.10+, and Foundry Local for the current desktop app dependency path

<Steps>
  <Step title="Install the CLI">
    ```bash theme={"system"}
    npm install -g taskbean
    ```

    Confirm the install:

    ```bash theme={"system"}
    bean --version
    ```
  </Step>

  <Step title="Record a task">
    ```bash theme={"system"}
    bean add "Document the authentication flow"
    bean list
    bean start 1
    bean done 1
    bean report --date today
    ```

    Taskbean resolves the current Workspace and rolls the task up to its Project.
  </Step>

  <Step title="Install the Agent Skill">
    For GitHub Copilot:

    ```bash theme={"system"}
    bean install --agent copilot
    ```

    To install for supported agent CLIs found on your `PATH`:

    ```bash theme={"system"}
    bean install --agent auto
    ```

    See [Agent Skill](/cli/agent-skill) for detection behavior, agent-specific installs, global installs, and sandbox configuration.
  </Step>

  <Step title="Launch the desktop app">
    From a clone of `taskbean/taskbean`, use the Windows launcher:

    ```powershell theme={"system"}
    cd app
    .\launch.ps1
    ```

    Or start the Python backend directly:

    ```bash theme={"system"}
    cd app/agent
    pip install -r requirements.txt
    python main.py
    ```

    Open `http://127.0.0.1:8275`. The app reads the same `~/.taskbean/taskbean.db` file as the CLI.
  </Step>
</Steps>

<Info>
  The first local model download can take several minutes. Check `GET /api/health` for process state and `GET /api/ready` for model readiness.
</Info>


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.