Agently

Active
GitHub Python Apache-2.0

Description

A GenAI application development framework that simplifies agent interaction with structured data and chained-calls syntax, using event-driven flow for complex logic.

Key Features

  • Structured output control with framework-guaranteed schemas, required field extraction, and retry validation
  • Runtime Skills system for discovering, installing, and executing MCP/script capabilities on demand
  • TriggerFlow event-driven workflows with fan-out, pause/resume, save/load, and sub-flow support
  • Action Runtime unifying local functions, MCP servers, shell/Python/Node helpers under one observable interface
  • Model switching without rewriting business logic through normalized provider contracts
  • Dynamic Task DAG execution converting LLM-generated plans into validated task graphs

Use Cases

💡 Building production AI service backends with reliable structured outputs and streaming
💡 Creating complex multi-step workflows with observable action execution and recovery
💡 Developing AI copilots and knowledge tools that survive model provider changes
💡 Building reusable Skills for cross-project AI capability sharing
💡 Deploying FastAPI-based AI services with DevTools observability and settings management

Quick Start

pip install agently

Related Projects