AGI Workforce
Back to Home

Documentation

Complete guides, API references, and feature documentation for AGI Workforce. Learn how to deploy autonomous AI agents across your desktop and web.

Getting Started

Quick start guide to set up and use AGI Workforce

Installation

Download and install the desktop application

Feature Documentation

Autonomous AI Agents

  • •Agent Planning & Execution
  • •Multi-step Task Orchestration
  • •Self-healing & Error Recovery
  • •Goal-based Reasoning

Multi-LLM Support

  • •OpenAI GPT-5 & GPT-4
  • •Anthropic Claude 4.5
  • •Google Gemini 3
  • •X.AI Grok 4.1
  • •DeepSeek V3.2
  • •Alibaba Qwen 3
  • •Local models via Ollama (Llama 3.3, etc.)

Desktop & Web Automation

  • •Browser automation with Playwright
  • •Screen capture & visual analysis
  • •File operations & document processing
  • •Cross-application workflows

Database & API Integration

  • •PostgreSQL, MySQL, SQLite support
  • •REST API calls
  • •Webhook support
  • •Custom tool development

Security & Privacy

AGI Workforce is built with security and privacy as top priorities:

  • Local-first execution - Your data stays on your device unless you explicitly allow cloud connections
  • Encrypted credential storage - API keys and sensitive data stored using OS keyring with AES-256-GCM encryption
  • Sandboxed agent environments - Agents run with strict permission controls and user approval for file/network access
  • No data sent to AGI Workforce servers - We don't collect or store your automation data or credentials
  • TLS 1.3 for all network connections - End-to-end encryption for all API communications

API Reference

AGI Workforce provides REST API endpoints for programmatic access:

POST/api/llm/v1/chat/completionsOpenAI-compatible chat completions endpoint
GET/api/llm/v1/modelsList available LLM models
GET/api/llm/v1/credits/balanceGet credit balance for authenticated user
POST/api/checkoutCreate Stripe checkout session
POST/api/portalAccess Stripe billing portal
POST/api/device/linkLink desktop device to account

OpenAI-Compatible API

The /api/llm/v1/chat/completions endpoint is fully compatible with OpenAI's API format, making it easy to integrate with existing tools and libraries.

Integration & Development

External Service Integration

  • •REST API calls with authentication
  • •Webhook support for event notifications
  • •OAuth 2.0 integration
  • •Custom tool development

Development Tools

  • •TypeScript SDK for custom integrations
  • •Rust backend for native extensions
  • •Visual workflow designer
  • •Built-in debugging tools

Need Help?

Can't find what you're looking for? We're here to help you get started.

Diagnostic ToolView FAQContact Support