License: FSL-1.1-MIT GitHub stars
Atelier is a full-stack AI agent orchestration platform for building, configuring, and managing AI agents with streaming chat, tool integrations, persistent memory, and collaborative workspaces.
📖 Read the Docs → — setup, usage, and everything else lives here.
An AI Workspace for creating and managing AI agents across multiple LLM providers (Google Gemini, OpenAI GPT, Anthropic Claude). Configure agents, connect external tools, schedule automated runs, orchestrate multi-agent workflows, and keep conversation history with semantic memory — all organized into collaborative workspaces for teams and individuals.
Highlights: multi-LLM support · team workspaces with RBAC · agent + skill builder · streaming chat · visual multi-agent orchestrator · MCP / database / file / Claude Code tools · cron scheduling · session & facts memory · Slack integration · Opik observability.
Full architecture, tech stack, feature reference, and project layout are in the docs.
See the User Guide for full setup instructions.
# Backend (FastAPI)
uv sync && uv run uvicorn app.main:app --reload
# Frontend (Next.js)
cd web && npm install && npm run devIf Atelier is useful to you, please star it on GitHub — it helps others find the project and keeps development going.
Want to donate, sponsor a feature, discuss commercial use, or reach out for anything else? Get in touch:
Atelier is source-available under the Functional Source License (FSL-1.1-MIT) — see LICENSE.md for full terms.
- ✅ Use, self-host, modify, contribute, and use internally (including at a company) — free of charge.
- ❌ Do not sell it or offer it as a competing commercial product/service (e.g. a hosted version of Atelier).
- 🕓 Two years after each release, that version converts to the MIT License.
It's source-available, not OSI "open source," because it restricts competing commercial use. For commercial licensing, email [email protected].