- Create comprehensive MCP.md documenting all available tools:
* Hugging Face Hub (models, datasets, papers, spaces, docs)
* Web search and fetch for research
* Code tools (Bash, Git, file ops)
* Claude Code (GSD) workflow agents
- Map MCP usage to specific phases:
* Phase 1: Model discovery (Mistral, Llama, quantized options)
* Phase 2: Safety research (sandboxing, verification papers)
* Phase 5: Conversation datasets and papers
* Phase 12: Voice visualization models and spaces
* Phase 13: Avatar generation tools and research
* Phase 14: Mobile inference frameworks and patterns
- Update config.json with MCP settings:
* Enable Hugging Face (mystiatech authenticated)
* Enable WebSearch for current practices
* Set default result limits
- Update PROJECT.md constraints to document MCP enablement
Research phases will leverage MCPs extensively for optimal
library/model selection, architecture patterns, and best practices.
- Update PROJECT.md: Add Android, visualizer, and avatar to v1
- Update REQUIREMENTS.md: 99 requirements across 15 phases (fresh slate)
- Add comprehensive README.md with setup, architecture, and usage
- Add PROGRESS.md for Discord forum sharing
- Add .gitignore for Python/.venv and project artifacts
- Note: All development via Claude Code/OpenCode workflow
- Note: Python deps managed via .venv virtual environment
Core value: Mai is a real collaborator, not a tool. She learns from you,
improves herself, has boundaries and opinions, and becomes more *her* over time.
v1 includes: Model interface, Safety, Resources, Memory, Conversation,
CLI, Self-Improvement, Approval, Personality, Discord, Offline, Voice
Visualization, Avatar, Android App, Device Sync.
