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

# Quickstart

> Connect a source and work towards your first client assessment.

## Before you begin

Sign in to [Alignr](https://app.alignr.io) with a workspace account. You need permission to manage integrations, clients and standards. Have the source system’s credentials or an authorised administrator available.

<Steps>
  <Step title="Connect a source">
    Open **Integrations**, choose your source and follow its connection form. Use the permissions requested by that connector. A successful connection establishes access; it does not mean a client has passed an assessment.

    [Learn about integrations](/guides/integrations).
  </Step>

  <Step title="Map your clients">
    Review the source’s companies or tenants. Link each record to the right existing client, or explicitly create a new client. Check names and identifiers before confirming the mapping.

    [Understand Organizations](/guides/organizations).
  </Step>

  <Step title="Review your standard">
    Open **Standards** and review the controls you want to apply. Check their parameters, evidence requirements and client-specific overrides before enabling them.
  </Step>

  <Step title="Review collected evidence">
    Check that collection has completed for the intended client. Look at source and observation time. Missing or stale evidence must not be interpreted as passing.
  </Step>

  <Step title="Review the results">
    Open the client’s alignment results. Investigate failures, distinguish missing data from unsupported coverage, and assign manual checks where needed. Review a supported remediation plan before approving a change.
  </Step>
</Steps>

## What success looks like

You can identify the client, the standard being assessed, and the evidence behind each result. An incomplete assessment is useful when it makes its gaps visible.

[Understand control statuses](/guides/control-status).

## A worked first assessment

Use a built-in identity standard as an example, after reviewing its controls and connecting a source that supplies the necessary observations.

1. Confirm the source's directory maps to the intended client.
2. Inspect an observed administrator's `has_role` fact and its source.
3. Inspect the same subject's `mfa_registered` observation.
4. Review the administrator MFA control and its selected population.
5. Read the result: an observed `false` can prove failure; a missing observation cannot prove a pass.

This checks registration, not every aspect of MFA enforcement. The [control walkthrough](/guides/control-conditions) explains the definition and the difference between passing, failing and unknown subjects.

## Learn the language as you go

You do not need to memorise technical names. A **predicate** names what is observed; a **fact** records a value for a subject; a **control** compares evidence with an expectation. Keep the [glossary](/guides/glossary) and [predicate reference](/guides/predicate-reference) available when reviewing unfamiliar results.