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

# Quickstart

> Connect one repository, run an explicit Entelligence review, and verify the result before enabling automation.

Connect one repository and run an explicit review. This path is easy to observe and does not depend on your workspace's automatic-review settings.

## Before you begin

You need:

* An Entelligence workspace
* Access to [**Connectors**](https://app.entelligence.ai/connectors)
* Permission to authorize or request access to a GitHub, GitLab, or Bitbucket repository
* A test pull request with a small, understandable change

<Note>
  If you are not a workspace administrator, send the [admin checklist](/get-started/admin-checklist) to the person who manages connectors and repository access.
</Note>

## 1. Connect source control

<Steps>
  <Step title="Open Connectors">
    In Entelligence, open [**Connectors**](https://app.entelligence.ai/connectors) and choose your Git provider.
  </Step>

  <Step title="Authorize a limited first scope">
    Select one test repository if your provider supports repository-level installation. You can expand access after verification.
  </Step>

  <Step title="Wait for synchronization">
    Return to **Code Review → Codebases → Repositories**. Confirm the repository appears. Use **Add Repo** or refresh the provider connection if it does not.
  </Step>
</Steps>

## 2. Enable the repository

Open [**Code Review**](https://app.entelligence.ai/code-review), select **Codebases**, then **Repositories**.

1. Search for the repository.
2. Enable **PR Review** for that repository.
3. Open **Show details** and review its current controls.
4. For a first test, keep the comment threshold narrow enough that results are easy to evaluate.
5. Save the repository settings.

<Warning>
  Workspace values shown in your account are not universal defaults. In particular, verify **PR Review Enabled**, **Auto Review on Every Commit**, comment style, and maximum inline comments before using the test repository for normal work.
</Warning>

## 3. Run a first review

Create or open a non-draft pull request. Request a review from the Entelligence bot using the command supported by your provider, or mention the bot and ask it to review.

You can also test locally:

```bash theme={null}
curl -fsSL https://app.entelligence.ai/install.sh | sh
entelligence auth login
cd /path/to/your/repository
entelligence status
entelligence review
```

The local command reviews the current Git changes and prints results in your terminal. It does not require posting comments to a pull request.

## 4. Verify success

A successful pull-request test has:

* An Entelligence summary or review on the expected pull request
* Findings attached to relevant lines when inline comments are enabled
* No access error from the Git provider
* The pull request visible under **Code Review → Codebases → Active PRs**

A successful CLI test has:

* A valid authenticated status
* A detected repository and diff
* Reviewed files and either findings or a clear no-findings result

## 5. Configure normal operation

After the first result is correct:

* Choose automatic or explicit review triggers.
* Choose inline comments or a single summary.
* Set the finding threshold and maximum inline comments.
* Add repository guidelines and Markdown context.
* Decide whether cross-repository context is appropriate for your data boundaries.
* Invite a small pilot team.

See [Code Review settings](/code-review/settings) for the complete control list.

## Next step

Learn [how to interpret findings](/code-review/findings), or follow the [Agent Insights rollout](/agent-insights/setup) to measure AI-assisted development.