🆕 Introducing Open Coworker — an open agent harness that runs tasks and automations on your machine. Desktop app, your own API keys, your files. → Get started in 2 minutes
Simple, unified tools for building with LLMs — from a drop-in chat client to a full agent harness. Three layers, each usable on its own.
A uniform client across 20+ providers using an OpenAI-compatible interface. Swap models with a string — no SDK juggling, no rewrites.
import aisuite as ai
client = ai.Client()
response = client.chat.completions.create(
model="anthropic:claude-sonnet-4-5",
messages=[{"role": "user", "content": "Why is the sky blue?"}],
)
print(response.choices[0].message.content)pip install aisuite # add provider extras, e.g. aisuite[anthropic]A lightweight Agent + Runner built on the client — tools, continuation
state, and tracing, without adopting a heavy framework. The agent is the
reusable definition; the runner owns execution.
import aisuite as ai
def get_weather(city: str) -> str:
"""Get the current weather for a city."""
return f"It's sunny in {city}."
agent = ai.Agent(
name="assistant",
model="openai:gpt-5.5",
instructions="Answer briefly. Use tools when they help.",
tools=[get_weather],
)
result = ai.Runner.run_sync(agent, "What's the weather in San Francisco?")
print(result.final_output)pip install aisuite[agents]An open agent harness for automating daily tasks and more — a desktop app where an agent works in your folders, uses connectors and tools, produces artifacts, and runs scheduled automations. Bring your own API keys; your files and keys stay on your machine.
- Multi-folder file access with per-folder read-only / read-write grants
- Connectors & tools — browser automation, integrations, MCP servers
- Artifacts — Markdown, images, PDF, CSV, spreadsheets, and Office files
- Automations — scheduled runs that continue as conversations
- Your models — OpenAI, Anthropic, Gemini, and local models via Ollama
Download for macOS / Windows · quickstart · docs
Each layer builds on the one below it, and each is usable on its own:
Open Coworker an agent harness for tasks & automations (app)
│ built on
Agent API Agent + Runner: tools, state, tracing (lib)
│ built on
Chat Completions one client, many providers (lib)
Start at whatever layer fits: call a model, wrap it in an agent, or run the whole harness.
libs/
aisuite-py/ the `aisuite` package — Chat Completions + Agent API
aisuite-js/ JavaScript/TypeScript port
apps/
opencoworker/ the Open Coworker harness (desktop app + server)
code-cli/ aisuite-code, the command-line coding agent
examples/ runnable examples per product
docs/ chat / agents / coworker
- Chat Completions — docs/chat
- Agent API — docs/agents
- Open Coworker — docs/coworker
We welcome contributions! See CONTRIBUTING.md to get started.