​
Supported Frameworks

RunAgent is designed to work with any AI agent framework. We provide first-class support for popular frameworks while allowing custom implementations.

LangGraph

Build stateful, multi-step agents with graph-based workflows

CrewAI

Create multi-agent systems with role-based collaboration

Agno

Develop agents with built-in memory and reasoning

Letta

Build conversational agents with long-term memory

LangChain

Orchestrate LLM chains into sophisticated workflows

LlamaIndex

Build context-aware document indices for LLM retrieval

​
Quick Start Examples

runagent init my-agent --framework langgraph
cd my-agent
# Edit agents.py with your LangGraph logic
runagent serve .

​
Framework Integration

All frameworks integrate with RunAgent through:
  1. Standardized Entrypoints: Define how to invoke your agent
  2. Configuration: Specify framework in runagent.config.json
  3. Environment Management: Handle API keys and settings
  4. Deployment: Same process regardless of framework

​
Best Practices by Framework

  • Keep graphs simple and readable
  • Use appropriate state persistence
  • Handle edge cases in transitions
  • Test each node independently
  • Define clear agent roles
  • Minimize agent dependencies
  • Use appropriate delegation patterns
  • Monitor inter-agent communication
  • Follow RunAgent conventions
  • Implement proper error handling
  • Document your approach
  • Consider open-sourcing if general-purpose

​
Migration Between Frameworks

RunAgent makes it easy to switch frameworks:
  1. Keep business logic separate from framework code
  2. Use consistent interfaces for inputs/outputs
  3. Test thoroughly after migration
  4. Update configuration in runagent.config.json

​
Community Frameworks

We’re always adding support for new frameworks. Coming soon:
  • AutoGen
  • BabyAGI
  • AgentGPT
  • Custom community frameworks

​
Contributing Framework Support

Want to add support for a new framework?
  1. Check our contribution guidelines
  2. Create a template for the framework
  3. Add documentation
  4. Submit a pull request

​
Next Steps

LangGraph Guide

Build your first LangGraph agent

CrewAI Guide

Create a multi-agent crew