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

# Quickstart

Five minutes from cold start to a Hermes Agent chatting with persistent memory.

## 1. Sign in

Open the app, click **Enter the Mirror**, and sign in with email or Google. MIRARI uses email-verified auth — you'll need to click the link in your inbox the first time.

## 2. Bind a Hermes Agent

Navigate to **Settings**. You'll see a row of presets:

| Preset         | Glyph | Endpoint                                         |
| -------------- | ----- | ------------------------------------------------ |
| Hermes · Local | ☿     | `http://localhost:11434/v1` (Ollama / llama.cpp) |
| Hermes · Cloud | ☿     | A hosted Hermes-family endpoint                  |
| OpenRouter     | ◈     | `https://openrouter.ai/api/v1`                   |
| Nous           | ✶     | Nous inference endpoint                          |
| Custom         | ⌘     | Anything that speaks OpenAI-compatible chat      |

Pick one, paste your API key (or leave blank for a local endpoint with no auth), give the binding a name, and click **Bind**.

> **Keys never leave your browser.** They are stored in your browser's local key vault, namespaced per connection id. The backend stores only the connection metadata (base URL, model name, label) — never the secret.

Click **Test** to send a one-token ping. A green ✓ means Hermes is awake.

## 3. Open the Console

Go to **Console**, pick your binding from the dropdown, and start a new conversation. Each message you send is persisted in the `conversations` + `messages` tables, so closing the tab does not lose context.

When the agent says something worth remembering, click **Promote to Memory** on the message. That single click creates a new node in the Memory Atlas with strength `0.5`.

## 4. Watch the Atlas grow

Open **Memory** to see every node — label, category, strength, source. Strong nodes (≥ 0.75) get injected as system context on future conversations. Weak nodes age out unless you re-promote them.

## 5. Reflect

Open **Mirror**, paste a few recent transcripts (or click *Import last conversation*), give the report a title, and run **Reflect**. You'll get back:

* **Strengths** — what the agent did well.
* **Weaknesses** — where it drifted.
* **Recommendations** — concrete fixes.
* **Suggested system prompt** — a redraft you can promote and A/B test.

## 6. Forge a skill

If a recommendation feels reusable, open **Skills** and click **Suggest Skill** with the relevant context. The agent proposes a skill (name, description, category, template prompt). Accept it → test it → promote it → it now ships as part of every relevant conversation's system context.

That's the whole loop: **Bind → Chat → Promote → Reflect → Forge.**


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://entermirari.gitbook.io/entermirari-docs/getting-started/quickstart.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.