BrainBlend-AI/atomic-agents
Building AI agents, atomically
This is a framework for software developers to build AI applications using a modular, 'LEGO-block' approach. It helps create AI systems that are predictable and maintainable by allowing developers to define clear inputs and outputs, and reuse components. Developers can build a wide range of AI tools, from simple chatbots to complex multi-step processes, ensuring consistent behavior.
5,758 stars. Actively maintained with 36 commits in the last 30 days. Available on PyPI.
Use this if you are a software developer who needs to build reliable, well-structured AI applications and integrate them into existing software systems, applying standard software engineering practices.
Not ideal if you are an end-user looking for a ready-to-use AI tool or a low-code/no-code solution.
Stars
5,758
Forks
475
Language
Python
License
MIT
Category
Last pushed
Feb 14, 2026
Commits (30d)
36
Dependencies
10
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/agents/BrainBlend-AI/atomic-agents"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Recent Releases
Related agents
openai/openai-agents-python
A lightweight, powerful framework for multi-agent workflows
openagents-org/openagents
OpenAgents - AI Agent Networks for Open Collaboration
vamplabAI/sgr-agent-core
Schema-Guided Reasoning (SGR) has agentic system design created by neuraldeep community
camel-ai/camel
🐫 CAMEL: The first and the best multi-agent framework. Finding the Scaling Law of Agents....
ServiceNow/AgentLab
AgentLab: An open-source framework for developing, testing, and benchmarking web agents on...