Features
Long-running agent execution with durability
Built-in tool use with custom tool creation
Model Context Protocol (MCP) support
Agent-to-Agent (A2A) protocol for interoperability
Streaming responses from agents
Tracing and monitoring for production
Skill system for modular agent capabilities
Cloud deployment via Celesto CLI
FastAPI integration for serving agents
Authorization in MCP servers
Gmail and Google Calendar integration guides
Open-source with extensible architecture
@ai_fn decorator for LLM-powered Python functions
@ai_classifier decorator for text classification
Structured data extraction via Pydantic models
Agent loops with tool calling and function calling
Built-in streaming (SSE) support
Async-first API for concurrent applications
Rate limiting and automatic retries
LLM response caching for performance
Concurrency control for workload management
CLI tool for monitoring and debugging
SQLite state store for persistence
OpenAI and Anthropic model support
Embeddings generation for semantic search
Local execution with no cloud dependency
Self-hosted deployable as a library
