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

# Tool Reference

> Every Autonomy MCP tool, grouped by domain.

The product MCP (`autonomy`) exposes **36 tools**, each a thin translation of one platform endpoint. They're grouped into four domains plus a guidance helper.

<CardGroup cols={2}>
  <Card title="Projects" icon="folder" href="/tools/projects">
    List, connect, and initialize projects.
  </Card>

  <Card title="Tasks" icon="play" href="/tools/tasks">
    Start, poll, iterate, review, and ship coding tasks.
  </Card>

  <Card title="Setup" icon="gear" href="/tools/setup">
    Configure commands, env vars, and validation.
  </Card>

  <Card title="Full Mode" icon="layer-group" href="/tools/full-mode">
    Proxy the app's API calls to your staging backend.
  </Card>
</CardGroup>

## Conventions

* **IDs** - most tools take a `project_id`; task tools also take a `task_id`. Get both from `list_projects` and `list_tasks` / `get_task`.
* **Background work** - `start_task`, `continue_task`, `complete_task`, and `update_task_env` return immediately and run in the background. Poll `get_task` to track them.
* **Read-only tools** are annotated as such and never mutate state (e.g. `list_projects`, `get_task`, `get_task_pr_diff`).

## Guidance helper

`load_autonomy_skill(name)` returns a built-in step-by-step playbook - no platform call. Names: `setup`, `full-mode`, `tasks`, `troubleshooting`. The server also exposes prompts: `/autonomy-getting-started`, `/autonomy-setup`, `/autonomy-new-task`, `/autonomy-full-mode`.

<Info>
  These reference pages are generated from the MCP server's tool definitions, so they stay in sync with the code. If a tool's signature changes, its page changes with it.
</Info>