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

# Quickstart Guide: Build Your First AI Agent with Layest

> Get up and running with Layest in under 5 minutes. Create a workspace, configure an agent, add a trigger, and run your first workflow.

This guide walks you through the fastest path to a working AI agent on Layest. By the end, you'll have a configured agent, a live trigger, and your first completed workflow run — all in under five minutes.

<Tip>
  The **Agent Library** contains pre-built agent templates for common use cases like customer support triage, data enrichment, and scheduled reporting. Browse it from **Agents → New Agent → Browse Library** to skip manual configuration and start from a working baseline instead.
</Tip>

<Steps>
  <Step title="Create or join a workspace">
    Go to [app.layest.com](https://app.layest.com) and sign up with your work email, or accept an existing invite from your team. When your workspace loads, you'll see the main dashboard.

    If you haven't set up your workspace yet — chosen a name, invited teammates, and configured your timezone — take five minutes to do that first.

    <Card title="Workspace Setup" icon="gear" href="/get-started/workspace-setup">
      Step-by-step workspace configuration, team invites, and dashboard orientation.
    </Card>
  </Step>

  <Step title="Create your first agent">
    From the left sidebar, click **Agents**, then click **New Agent** in the top-right corner.

    Fill in the agent basics:

    * **Name** — Give your agent a clear, descriptive name (e.g., `Support Triage Bot`).
    * **Instructions** — Write a system prompt that tells the agent its role, how it should behave, and any constraints it must follow. Be specific — the more context you provide, the more reliably the agent will act.
    * **Model** — Choose the underlying language model. If you're unsure, leave this as the workspace default.

    Click **Save agent** to create it. Your agent now exists but isn't connected to any capabilities or triggers yet.

    <Note>
      You can edit an agent's instructions, model, and capabilities at any time. Changes take effect on the next run — in-progress runs use the configuration they started with.
    </Note>

    For a deeper look at agent configuration options, see the [Agents guide](/agents/first-agent).
  </Step>

  <Step title="Add a capability or integration">
    Select your new agent and open the **Capabilities** tab. Here you can connect the tools your agent will use.

    To add an integration:

    1. Click **Add capability → Integration**.
    2. Browse the integration catalog and select a service (for example, Slack, Google Sheets, or a custom REST API).
    3. Authenticate with your credentials or API key.
    4. Choose which actions the agent is allowed to perform (read, write, or both).

    Save the capability. Your agent can now interact with that service as part of any workflow.

    <Tip>
      You can add multiple capabilities to a single agent. Start with one and add more as your workflow grows.
    </Tip>

    See the full [Integrations guide](/capabilities/integrations) for a catalog of available connectors and advanced configuration options.
  </Step>

  <Step title="Attach a trigger">
    A trigger defines when your agent runs. Open the **Triggers** tab on your agent and click **Add trigger**. Choose from:

    * **Manual** — Run the agent on demand from the dashboard or via the API. Great for testing.
    * **Schedule** — Set a cron expression or pick a simple interval (hourly, daily, weekly).
    * **Webhook** — Provide an external system with a unique URL; every POST to that URL fires the agent.
    * **Form submission** — Trigger the agent when a Layest form is submitted.

    For this quickstart, select **Manual** so you can fire the first run yourself without any external setup. Click **Save trigger**.

    <Warning>
      Webhook URLs are unique per trigger. Treat them like secrets — anyone with the URL can trigger your agent. Regenerate the URL from the trigger settings if it's ever exposed.
    </Warning>
  </Step>

  <Step title="Run your workflow and inspect the results">
    With your agent configured and a manual trigger in place, click **Run now** from the agent detail page. Layest will execute the workflow and open a live trace in the **Observability** panel on the right.

    You'll see each step as it executes:

    ```text theme={null}
    ✓ Trigger fired          0ms
    ✓ Agent instructions     12ms
    ✓ Capability: Slack      340ms
    ✓ Workflow complete      352ms
    ```

    Click any step to inspect its input, output, and latency. If something went wrong, the error message and stack trace appear inline — no digging through separate log files required.

    Once the run succeeds, you're done. You have a working Layest agent.

    Explore the [Observability overview](/observability/overview) to learn how to set up alerts, compare runs over time, and track performance metrics.
  </Step>
</Steps>

## Where to go from here

You've built a basic agent. Now explore what makes Layest powerful for teams at scale.

<CardGroup cols={2}>
  <Card title="Knowledge & Memory" icon="brain" href="/capabilities/knowledge/memory">
    Attach documents and vector memory so your agents always answer with the right context.
  </Card>

  <Card title="Integrations" icon="plug" href="/capabilities/integrations">
    Browse the full integration catalog and connect agents to your existing tools.
  </Card>

  <Card title="AI Assistant" icon="message-bot" href="/ai-assistant/chat">
    Use the built-in chat interface to query data and trigger agents conversationally.
  </Card>

  <Card title="Observability" icon="chart-line" href="/observability/overview">
    Monitor runs, set up alerts, and measure agent performance over time.
  </Card>
</CardGroup>