Welcome to the comprehensive documentation for the Antigravity Workspace Template — a production-grade starter kit for building autonomous AI agents on Google Antigravity.
- Quick Start Guide — Installation, local development, and first steps
- Project Philosophy — Core concepts and Artifact-First protocol
- Zero-Config Features — Automatic tool and context discovery
- MCP Integration — Connect to external tools and data sources
- Multi-Agent Swarm — Orchestrate specialist agents for complex tasks
- Development Roadmap — Current progress and future plans through Phase 9
Recursive summarization automatically compresses history—context limits are a thing of the past.
Generic ReAct pattern. Just register any Python function in src/tools/, and the Agent learns to use it automatically.
Optimized for Gemini 2.0 Flash's speed and function calling capabilities.
Call any OpenAI-compatible API via the built-in call_openai_chat tool (supports OpenAI, Azure, Ollama).
| Task | Documentation |
|---|---|
| Get started with the agent | Quick Start |
| Build a custom tool | Zero-Config Features |
| Connect to an MCP server | MCP Integration |
| Use multiple agents | Multi-Agent Swarm |
| Understand the architecture | Project Philosophy |
| See what's coming | Development Roadmap |
.
├── .antigravity/ # 🛸 Antigravity config/rules
├── .context/ # 📚 Knowledge base auto-injected
├── artifacts/ # 📂 Agent outputs (plans, logs, visuals)
├── src/ # 🧠 Agent source code
│ ├── agent.py # Main agent loop
│ ├── memory.py # JSON memory manager
│ ├── mcp_client.py # MCP integration
│ ├── swarm.py # Multi-agent orchestration
│ ├── agents/ # Specialist agents
│ │ ├── base_agent.py
│ │ ├── coder_agent.py
│ │ ├── reviewer_agent.py
│ │ └── researcher_agent.py
│ └── tools/ # Tool implementations
│ ├── demo_tool.py
│ └── mcp_tools.py
├── tests/ # ✅ Test suite
├── scripts/ # 🧪 Utility scripts
├── docker-compose.yml # Local dev stack
├── README.md # Main landing page
└── requirements.txt # Python dependencies
- Start with Quick Start
- Learn Zero-Config tool discovery
- Explore the swarm protocol
- Read Quick Start Docker section
- Check Development Roadmap Phase 9 (Enterprise Core)
- Configure MCP servers in MCP Integration
- Understand Project Philosophy
- Study Multi-Agent Swarm architecture
- Review Development Roadmap vision
- Read Project Philosophy
- Check Development Roadmap Phase 9
- Open an issue to propose ideas
- 🌐 Antigravity Official Docs
- 📘 MCP Protocol Specification
- 🐍 Python Documentation
- 🐳 Docker Documentation
- 🧪 Pytest Documentation
Q: Can I use this with OpenAI instead of Gemini?
A: Yes! Set OPENAI_BASE_URL and OPENAI_API_KEY in .env. See Quick Start for details.
Q: How do I add a custom tool?
A: Drop a Python file in src/tools/ with your functions. No registration needed! See Zero-Config Features.
Q: How do I deploy to production?
A: Use Docker! See Quick Start Docker section.
Q: Can I use multiple agents?
A: Yes! Use the swarm system. See Multi-Agent Swarm.
Q: How do I add context/knowledge?
A: Create files in .context/ directory. They're automatically loaded! See Zero-Config Features.
We welcome contributions at all levels:
Found a bug? Open an issue
Have an architectural idea? Ideas are contributions too!
Propose your thought
Ready to code? Check the Roadmap Phase 9 for open areas.
See a typo or unclear section? Submit a PR to improve the docs!
- 📖 Documentation: You're reading it! (or check README.md)
- 🐛 Bug Reports: GitHub Issues
- 💡 Feature Requests: GitHub Discussions
- 👥 Community: Star the repo to stay updated
- @devalexanderdaza — First contributor. Implemented demo tools, enhanced agent functionality, proposed the "Agent OS" roadmap, and completed MCP integration.
- @Subham-KRLX — Added dynamic tools and context loading (Fixes #4) and the multi-agent cluster protocol (Fixes #6).
This project is licensed under the MIT License. See LICENSE for details.
Latest Update: December 2025
Version: Phase 8 (MCP Integration) ✅
Happy building with Antigravity! 🚀