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

# Quickstart

> The easiest way to get started with Overmind.

Overmind has first class integration with cursor, claude code, opencode, and codex through our SKILL `/overmind` and MCP server, after signing up you can pretty much do everything from your coding agent.

> We've built the experince to be as easy and native to your coding agent as possible.

<Steps>
  <Step title="Get on console">
    Start by signing up at [console.overmindlab.ai](https://console.overmindlab.ai)
  </Step>

  <Step title="Copy the onboarding prompt">
    Pick your coding agent — **Cursor**, **Claude Code**, **OpenCode**, or **Codex** — and hit **Copy onboarding prompt**.
  </Step>

  <Step title="Paste the prompt into your coding agent">
    Paste the prompt into your coding agent (repo root as the workspace folder) and hit enter.

    * Overmind will now scan your repository for [AI Agents & Capabilities](/core/capabilities#the-scan), you can see them in the [Console](https://console.overmindlab.ai)
    * If you want any changes in how they are grouped, evals, etc, just ask your coding agent to do it.
    * Next step is to ensure your codebase is instrumented properly for [tracing using OTLP endpoint](/core/observability#the-otlp-endpoint), just run `/overmind ensure-tracing` in your coding agent.
  </Step>

  <Step title="Improve the agent with /overmind commands">
    The pushed capabilities are the starting point. From the same chat, each command drives one loop end to end:

    | Command                    | What it does                                            |
    | -------------------------- | ------------------------------------------------------- |
    | `/overmind ensure-tracing` | Inspect codebase and instrument it properly for tracing |
    | `/overmind dataset`        | Build, clean, upload, or export a dataset               |
    | `/overmind finetune`       | Fine-tune, deploy, and smoke-test a model               |
    | `/overmind optimise`       | Run prompt and code optimisation                        |
    | `/overmind backtest`       | Compare models in the repository                        |
  </Step>

  <Step title="Watch it land in the Console">
    Once tracing is in, open **Observability**.

    <Frame caption="default view is Task executions: one row per scored unit, with the task it bound to and its execution score.">
      <img src="https://mintcdn.com/overmind-b84ae13c/OG-4bZDDAnV78HJp/images/platform/observability-executions.jpg?fit=max&auto=format&n=OG-4bZDDAnV78HJp&q=85&s=88cd69b275e9332c0d19ce053beb6cc9" alt="Observability task executions view in the Overmind console" width="1456" height="821" data-path="images/platform/observability-executions.jpg" />
    </Frame>
  </Step>
</Steps>

## Next steps

<CardGroup cols={2}>
  <Card title="Structure your traces" href="/tracing/sdk-python">
    Bracket each run with `overmind.run()` and add `@tool` and `@observe` spans so traces mirror your agent's real control flow.
  </Card>

  <Card title="Read the pushed capabilities" href="/core/capabilities">
    What sync uploaded: capabilities, prompts, tools, behaviours, and the trajectory map.
  </Card>

  <Card title="Build a dataset from traces" href="/core/datasets">
    Select traces and land them as an eval or training dataset with provenance intact.
  </Card>

  <Card title="Work through MCP" href="/platform/mcp">
    The tool catalog your coding agent already has after onboarding.
  </Card>
</CardGroup>