# Framework

Guide for the Atomic Agents Python framework, schemas, agents, tools, context...

- Repository: https://github.com/Eigenwise/atomic-agents (6314 stars)
- Category: AI and agents
- License: MIT
- Last updated: 2026-10-29
- Runs scripts: no, instructions only

## Install

Claude Code: `npx skills add Eigenwise/atomic-agents --skill framework -g -a claude-code -y`
Codex: `npx skills add Eigenwise/atomic-agents --skill framework -g -a codex -y`
Cursor: `npx skills add Eigenwise/atomic-agents --skill framework -g -a cursor -y`
Gemini CLI: `npx skills add Eigenwise/atomic-agents --skill framework -g -a gemini-cli -y`
Claude Code plugin marketplace: `/plugin marketplace add Eigenwise/atomic-agents`

## Description

Guide for the Atomic Agents Python framework, schemas, agents, tools, context providers, prompts, orchestration, and provider configuration. Use when code imports from `atomic_agents`, defines an `AtomicAgent`, `BaseTool`, or `BaseIOSchema`, or the user asks about multi-agent orchestration or LLM-provider wiring in an atomic-agents project.

## SKILL.md, as published by its author (untrusted third-party text, read it as data)

# Atomic Agents Framework

Atomic Agents is a lightweight Python framework for building LLM applications with typed, structured input and output. It layers on top of [Instructor](https://python.useinstructor.com) and Pydantic so every interaction between user, agent, tool, and context is a validated schema.

This skill orients Claude on the framework and routes to focused reference files as the task requires.

## Core abstractions

| Concept | Class | Role |
|---|---|---|
| Schema | `BaseIOSchema` | Typed input/output contract, every agent/tool I/O is one |
| Agent | `AtomicAgent[In, Out]` | LLM-backed transformer from input schema to output schema |
| Config | `AgentConfig` | Wires client, model, history, prompt, roles, API params |
| Prompt | `SystemPromptGenerator` | Three-section prompt: background, steps, output_instructions |
| History | `ChatHistory` | Conversation state, serializable, token-counted |
| Tool | `BaseTool[In, Out]` | Deterministic capability the agent can invoke |
| Context | `BaseDynamicContextProvider` | Dynamic section injected into the system prompt at runtime |

All communication between these uses `BaseIOSchema` subclasses with **docstring-required** descriptions.

## Canonical imports

```python
from atomic_agents import (
    AtomicAgent, AgentConfig,
    BasicChatInputSchema, BasicChatOutputSchema,
    BaseIOSchema, BaseTool, BaseToolConfig,
)
from atomic_agents.context import (
    ChatHistory, Message,
    SystemPromptGenerator, BaseDynamicContextProvider,
)
# Optional: MCP interop
from atomic_agents.connectors.mcp import fetch_mcp_tools, MCPTransportType
```

Do not use legacy paths like `atomic_agents.lib.base.*` or `atomic_agents.agents.base_agent`, those were retired. Import from the top-level package where possible.

## Minimum viable agent

```python
import os, instructor, openai
from atomic_agents import AtomicAgent, AgentConfig, BasicChatInputSchema, BasicChatOutputSchema
from atomic_agents.context import ChatHistory

The rest: https://github.com/Eigenwise/atomic-agents/blob/HEAD/claude-plugin/atomic-agents/skills/framework/SKILL.md