AIGNE Framework
AI Genesis Native Engineering
An open source TypeScript framework for building AI agents. AIGNE provides composable workflow patterns, multi-model support, and deep integration with the Blocklet ecosystem — so you can focus on agent logic, not infrastructure plumbing.
Core Capabilities
Multi-Model Support
Work with OpenAI, Claude, Gemini, and other LLM providers interchangeably. Swap models without rewriting agent logic. Your workflows stay portable across providers.
Workflow Patterns
Built-in sequential, concurrent, routing, and handoff patterns for composing agents. Chain simple agents into sophisticated workflows using declarative configuration.
AFS Integration
Native integration with the Agentic File System for persistent agent state, structured data access, and cross-agent knowledge sharing.
MCP Protocol
First-class support for the Model Context Protocol, enabling standardized tool interfaces and interoperability with the broader AI tooling ecosystem.
How It Works
Define Agents
Create agents in TypeScript with natural language instructions, tool bindings, and model configuration. Each agent is a composable unit with a clear interface.
Compose Workflows
Wire agents together using built-in patterns: sequential pipelines, concurrent fan-out, intelligent routing, and agent-to-agent handoff. Mix and match as needed.
Deploy to Blocklet Server
Package your agent workflows as Blocklets and deploy to Blocklet Server for production-grade hosting, monitoring, and lifecycle management.
Start Building AI Agents
AIGNE is open source under the Elastic Search License. Clone the repo, follow the quickstart, and have your first agent running in minutes.