Onboarding V3 & 60-Second Quick Start

The Onboarding V3 & Quick Start Cockpit transforms the first-run experience from a tedious multi-step form into a rapid, visually rich onboarding flow. Designed to deliver a working, embodied AI companion in under 60 seconds, it offers two distinct onboarding paths: the unified 60-Second Quick Start Dashboard for users who want immediate embodiment, and the guided 6-Step Setup Wizard for meticulous customization.
60-Second Quick Start Cockpit

The Quick Start Cockpit provides a unified four-corner dashboard where users configure all primary faculties on a single cohesive canvas:
- Top Left (Vessel & Stage): Instant model selection across Live2D, VRM, Spine, and MMD avatars with live preview framing.
- Top Right (Consciousness & Brain): LLM provider selection (OpenAI, Anthropic, Gemini, DeepSeek, Ollama, RWKV WebGPU) and API key configuration with inline model discovery.
- Bottom Left (Soul & Persona): Character persona selection, SillyTavern card drag-and-drop, and nickname customization.
- Bottom Right (Voice & Timbre): Voice synthesis engine selection (Kokoro local WebGPU, PocketTTS, ElevenLabs, Fish Audio, Edge TTS) with one-click preview auditioning.
A prominent Launch AIRI Stage action performs atomic pre-flight validation and initializes the stage session in real time.
Guided Setup Wizard Walkthrough
For creators desiring deep configuration, the multi-step wizard guides the user through each foundational layer of AIRI embodiment:
1. Hardware & Environment Verification

The Welcome step automatically probes the host system:
- Detects GPU acceleration (WebGPU, Apple Metal, NVIDIA CUDA, Vulkan).
- Audits local memory and audio I/O capabilities.
- Offers returning users an instant Restore Backup (BYOS) option to recover configurations from S3, Cloudflare R2, or local archives without completing initial setup again.
2. Experience Archetypes

Sets the high-level operational posture of AIRI through tailored archetypes:
- Casual Companion: Warm conversational banter, spontaneous spoken pacing, and gentle desktop reactions.
- Productivity & Workspace Partner: Focused task execution, proactive reminders, code assistance, and active-window awareness.
- Gaming & Stream Backseat: Reactive banter, screen-watching commentary, and game state empathy.
- Creative Director & Roleplay: Unrestricted multi-actor acting cues, ComfyUI image manifestation, and branching scenario lore.
3. Soul & Persona Starter Roster

Features an 8-card visual starter roster spanning varied personality profiles, speaking styles, and default scenarios. Cards support:
- Live pronoun and username token interpolation (
$\rightarrow$ custom user handle). - One-click import of external SillyTavern CCv2/CCv3 PNG cards and native
.airibundles. - Seamless multi-actor identity tags (
<|ACTOR:...|>) preserving character lore.
4. Physical Vessel 3D Coverflow

An interactive 3D coverflow carousel rendering avatars in real time:
- Seamlessly cycles across Live2D cubism models, VRM 0.x/1.0 avatars, and Spine 2D skeletons.
- Live idle animations and physics settling directly on the setup canvas.
- Real-time model metadata inspection (author, polycount, expression blendshapes, and texture resolution).
5. Pre-Flight Honesty Matrix Finale

The final pre-flight screen audits operational integrity across all 6 cognitive faculties:
- Audio Input & Hearing: Mic active and speech recognition verified.
- Voice Synthesis Engine: Selected TTS provider and voice model ready.
- Memory Hierarchy: STMM, Long-Term Journal, and Dream Worker active.
- Consciousness & Reasoning: Primary LLM connection validated.
- Avatar Vessel & Stage Driver: 3D/2D renderer initialized with stage lighting.
- Autonomous Tools: Sandboxed local tools and MCP bridges verified.
A live Turn 0 Opening Dialogue Preview allows testing the companion's first greeting with full expression cue execution, clean actor badges (e.g. [ 👤 Aqua ]), and instant voice playback before launching onto the main stage.
Key Capabilities
- Unified 60-Second Quick Start Cockpit: Configure Vessel, Brain, Persona, and Voice on a single canvas.
- 6-Step Guided Customization Wizard: Deep step-by-step calibration from hardware probing to pre-flight matrix.
- Interactive 3D Vessel Coverflow: Real-time avatar preview across Live2D, VRM, Spine, and MMD.
- Archetype Presets: Instant system prompt alignment for companion, productivity, streaming, or roleplay.
- Pre-Flight Honesty Matrix: 6-faculty audit ensuring zero broken dependencies before stage launch.
- Turn 0 Dialogue Preview: Actor badge parsing, clean marker stripping, and live speech preview before deployment.