Onboarding wizard
A six-step wizard runs on first launch and after a factory reset. Each step writes its choice straight into settings, so closing the wizard part-way still leaves the app in a consistent state — you can finish setting things up from Settings later.
The steps
- 1. Welcome — a brief intro card. Click Get started to move on.
- 2. Name — an optional display name. Once set, it is available as the
{{USER_NAME}}template variable in any system prompt (custom instructions, Space instructions, snippets, per-chat overrides). - 3. Provider — the only required step. Pick Ollama and (optionally) install a suggested starter model, or add an OpenAI-compatible API key. Skipping with Esc or × here triggers a confirm dialog, because the app needs a provider before anything else can work.
- 4. Prompt — the global Custom instructions textarea. Whatever you type here becomes the system prompt for every new chat (unless a Space or per-chat override wins).
- 5. Features — toggle the default behaviour for Temporal awareness, Thinking, Web fetch and Low VRAM mode. The wizard ships with recommended defaults pre-selected (see below).
- 6. Final — closes the wizard. You land in a fresh chat with the model dropdown auto-opened so you can pick a model and start typing immediately.
Recommended defaults
On the Features step, Loach pre-selects the toggle states that make sense for most people:
- Temporal awareness — on. Models get a small “today is…” preamble so they can answer “what day is it?” correctly.
- Thinking — on. Reasoning-capable models stream their thought process into a collapsible block.
- Web fetch — on. URLs in your prompts are downloaded and inlined.
- Low VRAM — off. Only enable it if your machine is constrained.
Dismissing and resuming
Press Esc or click × on any step to close the wizard. Steps you have completed stay completed; the rest can be filled in later from Settings. The wizard does not re-appear on subsequent launches unless you run a factory reset (see Data management).