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

# AI Quickstart

> Set up your coding agent with Sail's docs MCP and workflow skills for migration and agent building.

Sail is designed and documented to work with coding agents, whether you prefer Claude Code, Codex, Cursor, OpenCode, or another.
It only takes a couple of minutes to set up your agents for success using the tools here.

## Overview

Start using Sail by providing your agent with Sail's docs and the patterns to
migrate existing apps or build observable agents on Sail.

<CardGroup cols={2}>
  <Card title="Sail docs MCP" icon="book-open" href="#step-1-give-your-agent-sail%E2%80%99s-docs">
    Connects your agent to Sail's documentation: the API, models, pricing, and
    SDK. Served live from these docs, so it stays current.
  </Card>

  <Card title="Sail workflow skills" icon="diagram-project" href="#step-2-give-your-agent-sail%E2%80%99s-agent-building-skills">
    Gives your agent the patterns to migrate existing apps and to build,
    instrument, and debug observable agents on Sail.
  </Card>
</CardGroup>

## Step 1: Give your agent Sail's docs

Connect your tool to the Sail docs MCP server at
`https://docs.sailresearch.com/mcp`. It gives your agent a live connection to
these docs, so answers stay current:

* **Claude Code:** `claude mcp add --transport http sail-docs https://docs.sailresearch.com/mcp`
* **Claude Desktop:** Settings → Connectors → Add custom connector, then paste the URL.
* **ChatGPT:** Settings → Apps & Connectors → turn on Developer mode → Add new connector, then paste the URL.
* **Cursor:** Settings → MCP, then add the URL.

See the [MCP server guide](/mcp-server) for more clients and example
questions.

**Prefer not to install anything?** Every docs page has a copy button at the
top if you want to paste content directly into your agent interface. Or point
your tool at [llms.txt](https://docs.sailresearch.com/llms.txt) or
[llms-full.txt](https://docs.sailresearch.com/llms-full.txt).

## Step 2: Give your agent Sail's workflow skills

The canonical package is
[sailresearchco/sail-skills](https://github.com/sailresearchco/sail-skills).
It packages skills for migrating existing apps to Sail and for building
observable agents on Sail. An agent run is recorded as a Voyage: a trace of
named agents, spans, and events, with every model call and Sailbox command
attributed to the right step.

| Skill                        | Use it when                                                                                                                                                                            |
| ---------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| `sail-migrate`               | Migrate an existing app, agent, or workflow to Sail. Inference call sites and third-party sandboxes move to Sail with the same behavior, with an optional before and after comparison. |
| `sail-voyage`                | Build or instrument any Voyage, the entrypoint. Agents, spans, events, Sailbox exec attribution, terminal lifecycle. Includes a minimal runnable example.                              |
| `sail-inference-with-voyage` | Attribute Sail inference model calls to the active agent and span.                                                                                                                     |
| `sail-voyage-debugging`      | A Voyage ran but renders wrong in the dashboard, from symptom to cause to fix.                                                                                                         |
| `sail-fanout-policy`         | Offload heavy coding or analysis subtasks to Sail workers through the `sail_delegate` and `sail_fanout` tools, and apply the diffs the workers return. Claude Code only for now.       |

Install for your agent:

<CodeGroup>
  ```text Claude Code theme={null}
  /plugin marketplace add sailresearchco/sail-skills
  /plugin install sail@sail
  ```

  ```bash Codex theme={null}
  codex plugin marketplace add sailresearchco/sail-skills
  codex plugin add sail@sail
  ```

  ```text Cursor / others theme={null}
  # Copy the skills/ folders into your project, or download the repo ZIP:
  # https://github.com/sailresearchco/sail-skills
  ```
</CodeGroup>

In Claude Code the skills load automatically when relevant. Invoke one directly
as `/sail:sail-migrate` or `/sail:sail-voyage`.

<Note>
  Before you run generated SDK code, install `sail` and export `SAIL_API_KEY`.
  Do not rely on `sail auth login` as the SDK auth path for a plain Python agent
  process.
</Note>

## Next steps

**Already have an agent or workflow?** Install the skills above, then ask
your coding agent to "Migrate this app to Sail". The `sail-migrate` skill
guides the whole migration, including moving sandboxed execution to Sail.
Prefer not to install anything? Copy the migration prompt from
[Migrate to Sail](/migrate#copy-migration-prompt) instead.

**Starting fresh?** Use the `sail-voyage` skill to build your own observable AI
workflow. Ask your coding agent:

```text theme={null}
Build a small background research agent on Sail and make the whole run show
up as a trace I can open in the dashboard, each step, the model calls it
makes, and the sandbox commands it runs, attributed to the right part of the
workflow.
```

The `sail-voyage` skill takes it from there.

**Or** make a plain inference call first. See the [inference Quickstart](/quickstart).