diff --git a/docs/user-guide/quickstart/overview.md b/docs/user-guide/quickstart/overview.md index fd174e9d..88daab13 100644 --- a/docs/user-guide/quickstart/overview.md +++ b/docs/user-guide/quickstart/overview.md @@ -4,7 +4,7 @@ The Strands Agents SDK empowers developers to quickly build, manage, evaluate an ## :material-language-python: **Python Quickstart** -Create your first Python Strands agent! +Create your first Python Strands agent with full feature access! [**→ Start with Python**](python.md) @@ -13,8 +13,71 @@ Create your first Python Strands agent! ## :material-language-typescript: **TypeScript Quickstart** !!! info "Experimental SDK" - The TypeScript SDK is experimental + The TypeScript SDK is experimental with limited feature coverage compared to Python. Create your first TypeScript Strands agent! [**→ Start with TypeScript**](typescript.md) + +--- + + +## Language Support + +Strands Agents SDK is available in both Python and TypeScript. The Python SDK is mature and production-ready with comprehensive feature coverage. The TypeScript SDK is experimental and focuses on core agent functionality. + +### Feature Availability Matrix + +The table below shows feature availability across both SDKs. Python is the primary SDK with complete feature coverage, while TypeScript provides essential agent capabilities with ongoing development. + +| Feature Category | Feature | Python | TypeScript | Notes | +|------------------|---------|--------|------------|-------| +| **Core Agent** | Basic agent creation | ✅ | ✅ | | +| | Agent invocation | ✅ | ✅ | | +| | Streaming responses | ✅ | ✅ | | +| | Agent state management | ✅ | ✅ | | +| | Structured output | ✅ | ❌ | TypeScript support planned | +| **Model Providers** | Amazon Bedrock | ✅ | ✅ | | +| | OpenAI | ✅ | ✅ | | +| | Anthropic | ✅ | ❌ | | +| | Google Gemini | ✅ | ❌ | | +| | LiteLLM | ✅ | ❌ | | +| | Ollama | ✅ | ❌ | | +| | Mistral | ✅ | ❌ | | +| | Llama API | ✅ | ❌ | | +| | SageMaker | ✅ | ❌ | | +| | Writer | ✅ | ❌ | | +| | Custom providers | ✅ | ❌ | | +| **Tools & Extensions** | Custom function tools | ✅ | ✅ | | +| | MCP (Model Context Protocol) | ✅ | ✅ | | +| | Built-in vended tools | Community package | 4 built-in tools | bash, file_editor, http_request, notebook | +| | Tool executors (sequential/concurrent) | ✅ | ❌ | | +| | Hot tool reloading | ✅ | ❌ | | +| | Automatic tool loading | ✅ | ❌ | | +| **Conversation Management** | Null conversation manager | ✅ | ✅ | | +| | Sliding window manager | ✅ | ✅ | | +| | Summarizing conversation manager | ✅ | ❌ | | +| **Multi-Agent Patterns** | Agent swarm | ✅ | ❌ | | +| | Agent workflows | ✅ | ❌ | | +| | Agent graphs | ✅ | ❌ | | +| | Agent-to-agent communication | ✅ | ❌ | | +| | Agents as tools | ✅ | ❌ | | +| **Session Management** | File session manager | ✅ | ❌ | | +| | S3 session manager | ✅ | ❌ | | +| | Repository session manager | ✅ | ❌ | | +| | Custom session managers | ✅ | ❌ | | +| **Observability** | OpenTelemetry traces | ✅ | ❌ | | +| | Metrics collection | ✅ | ❌ | | +| | Custom telemetry | ✅ | ❌ | | +| **Hooks & Lifecycle** | Agent lifecycle hooks | ✅ | ✅ | | +| | Custom hook providers | ✅ | ✅ | | +| | Multi-agent hooks | ✅ | ❌ | Experimental in Python | +| **Experimental Features** | Bidirectional streaming | ✅ | ❌ | Real-time voice/audio agents | +| | Agent steering | ✅ | ❌ | Dynamic behavior modification | +| | Agent configuration | ✅ | ❌ | Runtime configuration management | +| **Interrupts & Control** | Agent interrupts | ✅ | ❌ | Graceful stopping and resumption | +| | Context overflow handling | ✅ | ❌ | | +| **Development Tools** | Debug logging | ✅ | ✅ | | +| | Agent builder | ✅ | ❌ | AI-powered agent generation | +| | Community tools package | ✅ | ❌ | 30+ production-ready tools | +