KodeAgent
KodeAgent is a frameworkless, minimalistic approach to building AI agents. Written in approximately 3,000 lines of pure Python, it is designed to serve as a robust reasoning core inside larger systems rather than being a heavy, all-encompassing platform.
Cost / License
- Free
- Open Source (Apache-2.0)
Platforms
- Mac
- Windows
- Linux
- Python
- Docker
KodeAgent
Features
Properties
- Lightweight
Features
- Subtasks
- Agentic AI
- Debugging
- Planning
Tags
- AI Agent
- azure-openai
- create-ai-agents
- Automation
- react-agent
- AI
- gemini-api
- agent
- codeagent
- litellm
- tasks
- e2b
- agentic-ai-development
- agentic-ai-automation
- agentic-framework
- Python
- ai-agent-platform
KodeAgent News & Activities
Recent activities
KodeAgent information
Developed by
barun-sahaLicensing
Open Source (Apache-2.0) and Free product.Written in
Alternatives
0 alternatives listedSupported Languages
- English
AlternativeTo Categories
Development, Office & Productivity, AI Tools & ServicesGitHub repository
- 16 Stars
- 1 Forks
- 20 Open Issues
- Updated
What is KodeAgent?
KodeAgent is a frameworkless, minimalistic approach to building AI agents. Written in approximately 3,000 lines of pure Python, it is designed to serve as a robust reasoning core inside larger systems rather than being a heavy, all-encompassing platform.
Adhering to the Unix Philosophy—"do one thing well and integrate seamlessly"—KodeAgent is lightweight, scalable, and ideal for serverless/application integration. It provides a "lightweight glass box" experience, allowing developers to read and debug the entire source without fighting opaque abstraction layers.
Key Features:
?LLM Agnostic: Built on LiteLLM, easily swap between Gemini, OpenAI, Claude, and others. ?Dual Paradigm Support: Out-of-the-box support for both ReAct and CodeAct agent patterns. ?Internal Observer: Features an "Auto-Correction" mechanism that monitors progress, detects loops, and provides feedback to keep the agent on track. ?Code Execution: Supports executing generated code in local 'host' environments or secure E2B sandboxes. ?Built-in Tools: Includes tools for web searching, Wikipedia, arXiv, audio transcription, image generation, and more. ?Observability: Native integration with Langfuse and LangSmith for detailed tracing.