> For the complete documentation index, see [llms.txt](https://docs.perception.cx/perception/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.perception.cx/perception/enma-ai/getting-started.md).

# Getting Started

## Install

1. [Download the installer](https://perception.cx/percapi/enma/harness/pub/enma_harness.exe) and run it. It installs per-user, no admin prompt.
2. Windows SmartScreen may warn about an unknown publisher. This is expected for an unsigned app: click "More info", then "Run anyway".

## First run

Enma asks for three things:

1. **Endpoint.** The base URL of any OpenAI-compatible server, ending in `/v1`. Examples: `http://localhost:1234/v1` (LM Studio), `http://localhost:11434/v1` (Ollama), `https://openrouter.ai/api/v1`.
2. **API key.** Only if your endpoint needs one. Local servers usually don't.
3. **Model.** Use the **Fetch models** button to list what your endpoint serves and pick one. Your key never leaves your machine.

That's it. You're chatting in under a minute.

{% hint style="info" %}
No cloud account? A free local model through LM Studio or Ollama works fine.
{% endhint %}

## Your first chat

1. Click **New chat** and pick a project folder.
2. Ask for something real: "add a dark mode toggle to this app" or "find and fix the bug in the login flow".
3. Watch the agents work. Every file change shows an inline diff, and you approve tool actions unless you loosen the permission mode.

## Updating

Installed copies check for updates automatically and download them in the background. When one is ready you get a "Restart to update" button in the title bar, with the changelog shown before you restart. Updates never interrupt a running agent. You can also check manually in Settings.