First-day walkthrough. Takes about 10 minutes end-to-end and leaves you with a working alpi — a model picked, a workspace pinned, messaging over Telegram optional, and the gateway running as a background service.
For concepts behind any step, see ARCHITECTURE.md. For the security posture, see SECURITY.md. Why alpi is built the way it is: see the Why section in the README.
1. Install
uv tool install /path/to/alpi
Needs Python ≥ 3.10. Optional: playwright install chromium if you want the browser tool (~200 MB download).
Check it works:
alpi --version
2. Pick a model
The fresh profile ships without a default model — you choose your provider and your model. Open the setup wizard:
alpi setup
Pick Model / Provider, choose a provider, paste the API key when prompted, pick a model. For recommendations see docs/MODELS.md — if you want a single sensible choice, Claude Sonnet 4.6 (anthropic/claude-sonnet-4-6) is the pragmatic daily driver. If you want fully local (no cloud), install Ollama first and pick Ollama in the wizard.
3. Pin a workspace
Same wizard, pick Workspace. Point it at the directory alpi is allowed to read and write — typically the project you're working on. Without a workspace, alpi falls back to the current working directory at launch with a warning; that's fine for trying things, not for real use.
4. Send your first message
alpi
Type hola and press Enter. The TUI streams the model's reply. Useful first commands to try inside the TUI:
/help— list every slash command./memory— show what alpi remembers about you (USER.md,MEMORY.md,AGENT.md)./skills— list installed skills./status— session snapshot (turns, tokens, cost)./new— start a fresh session.
Leave it with Ctrl-C or /exit.
5. Resume where you left off
alpi -c
-c / --continue rehydrates the last session. To make this the default behaviour without typing -c every time, enable tui.auto_resume in ~/.alpi/config.yaml (or set it via alpi setup).
6. Connect a gateway (optional)
If you want alpi to answer you on Telegram / IMAP / Gmail, run the wizard again:
alpi setup → Gateways
Pick the channel, follow the prompts (the wizard knows where to point you for tokens and credentials), then install the gateway as a background service:
alpi setup → Gateway service → Install
From now on alpi answers you over the channel 24/7 — on boot, on crash-restart, on every inbound. Messages from the gateway share per-chat session threads with your TUI, so a conversation started on Telegram can continue in the terminal.
7. Add a second profile (optional)
If you want a work profile that's completely isolated from your personal one (different API keys, different memory, different Telegram bot):
alpi profile create work
alpi -p work setup
Everything is per-profile: home directory, sessions, memory, skills, keys, peers. See docs/PROFILES.md for the full model.
8. Install ALP service (optional)
If you plan to link profiles or machines with other alpis, install the Alpi Link Protocol listener as a service and exchange pubkeys:
alpi setup → ALP service → Install
alpi setup → Peers
See docs/DEPLOYMENTS.md for topologies (laptop + home server, multi-device, team, enterprise).
9. Check everything is healthy
alpi doctor
Runs live checks: model reachable, gateways logged in, MCP servers handshaking, services alive, ALP socket listening, peers reachable. Green = you're done.
Next steps
- docs/SKILLS.md — teach alpi reusable recipes (skills are live-by-default, scanner-gated, per-profile).
- docs/OPERATIONS.md — logs, services, upgrades, backups.
- docs/SECURITY.md — approval gate, sandbox, and how alpi protects you from the LLM.
- docs/DEPLOYMENTS.md — multi-alpi topologies from personal to enterprise.