Summary

Foundry offers a ready-to-use AI Digital Company Platform, simplifying the deployment of autonomous AI teams. It transforms strategic goals into actionable business operations with minimal user intervention, ideal for users seeking to leverage AI companies without extensive custom coding or agent framework management.

How It Works

Foundry functions as a pre-built AI company, distinct from agent frameworks requiring manual orchestration. Its core is a multi-agent system with defined roles (CEO, Department Head, Employee) collaborating via real-time chat. Key innovations include an "Autonomous Heartbeat Loop" where a CEO Agent independently plans and executes tasks, and a "Collaboration Pipeline" using LangGraph for intent classification, task execution, and human-in-the-loop approvals. A layered RAG-enabled memory system ensures context retention and learning.

Quick Start & Requirements

• OS: Windows 10+, macOS 12+, Ubuntu 20.04+.
• Recommended: 4+ cores CPU, 16 GB RAM, 10 GB+ free SSD, Docker (latest), Node.js (22 LTS), pnpm (latest).
• Setup: One-command: git clone https://github.com/axislab-top/Foundry.git && cd Foundry && pnpm setup:dev.
• First Launch: 5-10 minutes (Docker image downloads ~2 GB); subsequent starts ~30 seconds.
• Resources: Official Quick Start, Features, Architecture, Comparison, Docs, Live Demo, 中文文档.

Highlighted Details

• One-Click Company Creation: Auto-generates org structures with drag-and-drop customization and industry templates.
• Multi-Agent Collaboration: Employs CEO, Department Head, and Employee agents with defined roles.
• Real-time Group Chat: Features dynamic conversations, human-in-the-loop approval, and progress notifications.
• Layered Memory System: 3-tier (Company/Department/Agent) RAG-enabled memory for persistent learning.
• Autonomous Operation: Driven by a CEO Agent's ~30-minute "Autonomous Heartbeat Loop" for continuous task management.
• Cost & Governance: Includes token tracking, budget controls, model routing, and audit logging.
• Core Differentiators: "Autonomous Heartbeat Loop" and "Collaboration Pipeline" enable self-improving, autonomous operations.

Maintenance & Community

The project provides a CONTRIBUTING.md file. Specific details on active maintenance, community channels (Discord/Slack), sponsorships, or a public roadmap are not explicitly detailed in the provided README text.

Licensing & Compatibility

• License: GPL-3.0.
• Commercial Use: Permitted, provided derivative works are also open-sourced under GPL-3.0. Direct commercial licensing inquiries are welcomed.

Limitations & Caveats

The GPL-3.0 license imposes copyleft requirements, necessitating that modifications or derivative works be open-sourced, impacting closed-source commercial integrations. Users must configure external LLM API keys and company details via the Admin Panel for AI agents to function. Initial setup requires Docker image downloads, and recommended system specifications are advised for optimal performance.