🚧 Limitations & Future Plans
Kaizen Agent is evolving rapidly, but it's important to understand what it can and can't do today — and where it's going.
Current Progress
| Feature | Python | LlamaIndex | LangChain | CrewAI | Mastra |
|---|---|---|---|---|---|
| Single Input + Single Output | ✅ | ✅ | ❌ | ❌ | ✅ |
| Multiple Inputs | ✅ | ✅ | ❌ | ❌ | ✅ |
| Async Support | ✅ | ✅ | ❌ | ❌ | ✅ |
| Track Values | ❌ | ❌ | ❌ | ❌ | ❌ |
| Browser Agent | ❌ | ❌ | ❌ | ❌ | ❌ |
| Advanced Tools | ❌ | ❌ | ❌ | ❌ | ❌ |
| Non-text Input/Output | ❌ | ❌ | ❌ | ❌ | ❌ |
| Multi-turn Conversations | ❌ | ❌ | ❌ | ❌ | ❌ |
| Multi-agents | ❌ | ❌ | ❌ | ❌ | ❌ |
Legend:
- ✅ = Fully Supported
- ❌ = Not Yet Supported
- 🔄 = In Development
Our long-term vision is to create an AI development teammate that accelerates LLM development automatically, so humans can focus on building features — and Kaizen Agent handles everything else to reach production-level reliability.