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

# Get started

> Install Vitalog, connect a client and record your first observation.

Vitalog is a self-hosted ledger for one person. There is no shared hosted account or team workspace. Each installation has its own PostgreSQL database, dashboard and client credentials.

<Steps>
  <Step title="Install Vitalog">
    Follow [Installation](/installation) to start the API, web app and PostgreSQL with Docker Compose. Configure a primary API key, database password, root sign-in credentials and timezone before starting.

    Use your own UI and API origins. Examples such as `https://vitalog.example.com` and `https://vitalog-api.example.com` are placeholders. This documentation site does not host your health records.
  </Step>

  <Step title="Sign in and connect a client">
    Open your installation's `/login` page and sign in with its configured root email and password. In **Account Settings → MCP Guide**, choose your client and copy its configuration. Approve the OAuth request to connect to the API's `/mcp` endpoint.

    For a REST script or Bearer-based MCP client, create a key in **Account Settings → API Keys**. Name, Permissions and Expires after are required. Choose Read-only, Edit or Administrative permissions, and 30 days, 90 days, 1 year or Never. Use Edit for logging observations. Copy the token when it is shown; it cannot be retrieved later.

    See [MCP setup](/mcp-guide) and [API keys](/api-keys).
  </Step>

  <Step title="Record an observation">
    Ask your MCP client to record a supplied observation, such as a 250 mL glass of water, with its actual date and timezone. The client can discover input schemas with `health_get_catalog` before logging.

    For a complete REST example, see [Log an observation](/logging). Every write needs an idempotency key. Reuse that key and the same arguments when retrying an uncertain response.
  </Step>

  <Step title="Review the result">
    Open **Daily View**, select the observation's date and expand its log. **Weight Management** shows weigh-ins and a 30-day graph when weight observations exist.

    Set your own [goals](/goals) through MCP or REST to add progress bars. The dashboard views health records; it does not create or edit them.
  </Step>
</Steps>

<Note>
  All screenshots and examples use synthetic data. Example goals illustrate the
  software and are not recommended health targets.
</Note>


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