# Eigenwise/atomic-agents

Building AI agents, atomically

Repository: https://github.com/Eigenwise/atomic-agents
Canonical: https://ross.abutalabs.com/products/atomic-agents
Homepage: https://eigenwise.github.io/atomic-agents/
Language: Python
License: MIT
License Family: permissive
Topics: ai, artificial-intelligence, large-language-model, large-language-models, llms, openai, openai-api
Last push: 2026-08-24T12:20:08+00:00

## Health v2 (maintenance only)
Score: 91/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 99, release rhythm 99, longevity 58
- inputs: {"age_days": 821, "days_push": 9, "days_rel": 9, "gap_med": 10.5, "n_releases_24m": 51}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 6202, forks 535 (observed 2026-08-28T04:09:39.804081+00:00)

## What it is
Atomic Agents is a lightweight, modular Python framework for building agentic AI pipelines and applications from small, composable components (agents, tools, context providers). It is built on Instructor and Pydantic, emphasizing schema-validated, predictable outputs and strong developer experience.

## Use cases
- build ai agents in python
- create multi-agent orchestration pipelines
- build a rag chatbot
- add tools and memory to an llm application
- get structured json outputs from llms
- build an mcp agent
- summarize youtube videos with ai

## When to choose
- you want a lightweight, composable alternative to heavyweight agent frameworks
- you value Pydantic-based schema validation and predictable agent outputs
- you need modular pipelines with hooks, tools, and context providers
- you prefer software-engineering-first patterns for LLM apps

## When to avoid
- you need a batteries-included all-in-one agent platform with a UI
- your stack is not Python
- you need built-in vector database or hosting rather than library-level composition

## Facets
- artifact type: framework
- maturity: active
- function: agent-framework, llm-inference, rag, prompt-engineering, mcp, chatbot, serialization
- domain: artificial-intelligence, large-language-models, developer-tools
- platform: python, cross-platform
- tags: pydantic, instructor, modular, multi-agent, structured-outputs, openai, ai-agents, retrieval-augmented-generation

## Member repositories
- Eigenwise/atomic-agents (main) score 91

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:39.804081+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:47:19.771364+00:00, confidence not recorded.
  - readme: https://github.com/Eigenwise/atomic-agents (fetched 2026-08-28T04:09:39.804081+00:00, sha f78009eb1858)
  - homepage: https://eigenwise.github.io/atomic-agents/ (fetched 2026-08-29T08:43:46.200752+00:00, sha e4b012cf43f9)
  - registry_pypi: https://pypi.org/pypi/atomic-agents/json (fetched 2026-08-29T08:43:46.209849+00:00, sha 0e6f0f0abd70)
- Data as of 2026-08-30T08:39:29.467469+00:00.