Open Source · Self-Hosted · Privacy-First

Your AI Agents,
Your Infrastructure

Name: ForgeAI
Author: ForgeAI

ForgeAI is a powerful open-source platform for deploying autonomous AI agents with multi-LLM routing, 13 built-in tools, 7 messaging channels, RAG, voice, and enterprise-grade security — all self-hosted.

Quick Start View on GitHub

$ git clone https://github.com/forgeai-dev/ForgeAI && pnpm install && pnpm forge start

LLM Providers

Built-in Tools

Channels

Dashboard Pages

API Tests

Features

Everything you need to run
autonomous AI agents

Multi-LLM Router

Route between 9 providers OpenAI, Anthropic, Google, Mistral, Groq, DeepSeek, xAI, Cohere, and Ollama (local). Automatic failover with circuit breaker.

13 Built-in Tools

Web browsing, search, Puppeteer (21 actions), shell, code sandbox, image generation (DALL-E 3, Leonardo AI, SD), file I/O, calculator, and more.

7 Messaging Channels

WhatsApp, Telegram, Discord, Slack, Microsoft Teams, Google Chat, and WebChat. One agent, every platform.

RAG Engine

Retrieval-Augmented Generation with TF-IDF or OpenAI embeddings, file upload (PDF, TXT, MD, code), persistence, and runtime config API.

Enterprise Security

7 security modules encrypted vault, RBAC, rate limiting, prompt guard, input sanitizer, 2FA, and tamper-proof audit logging with hash chains.

Voice & Vision

Speech-to-text, text-to-speech, image analysis, and image generation. Full multimodal support across all channels.

Plugin System & MCP

Extensible plugin architecture with a built-in store. Model Context Protocol (MCP) client for connecting to external tool servers.

Workflows & Autopilot

YAML-based workflow engine with cron scheduling, conditional steps, and an autopilot mode for fully autonomous agent operation.

Integrations

GitHub, Gmail, Google Calendar, Notion, and RSS feeds built-in. OAuth2 support for connecting to any service.

Architecture

Built for scale,
designed for developers

Modular Monorepo

8 packages: core, agent, tools, plugins, channels, security, workflows, shared. Clean boundaries, zero circular deps.

TypeScript + Fastify

100% TypeScript. Fastify HTTP gateway with WebSocket real-time. Sub-millisecond routing.

Security-First

AES-256-GCM encrypted vault, PBKDF2 key derivation, HMAC integrity checks, tamper-proof audit trail.

Docker Ready

Docker Compose with MySQL, Redis, and gateway. One command to deploy your entire AI infrastructure.

docker-compose.yml

version: '3.8'
services:
  gateway:
    image: forgeai/gateway:latest
    ports:
      - "18800:18800"
    environment:
      - OPENAI_API_KEY=sk-...
      - ANTHROPIC_API_KEY=sk-ant-...
      - LLM_PROVIDER=openai
    volumes:
      - ./data:/app/.forgeai
    depends_on:
      - db

  db:
    image: mysql:8
    environment:
      - MYSQL_DATABASE=forgeai

Providers

One API,
9 LLM providers

Switch between providers with a single config change. Automatic failover ensures your agents never go down.

OpenAI GPT-4o, o1, o3

Anthropic Claude Opus, Sonnet

Google Gemini Pro, Ultra

Mistral Large, Codestral

Groq Ultra-fast inference

DeepSeek V3, R1

xAI Grok

Cohere Command R+

Ollama (Local) 100% private, no API key

Dashboard

Beautiful control center
18 pages, real-time

Manage agents, monitor usage, configure tools, upload RAG documents, and view audit logs all from one dashboard.

ForgeAI Dashboard http://localhost:18800

Overview

Messages

12,847

Tokens

2.4M

Tools Used

1,293

Uptime

99.9%

Overview Chat Tools Usage Plugins Channels Agents Workspace Gmail Memory RAG API Keys Webhooks Calendar Voice Audit Log Settings + more

Your AI Agents, Your Infrastructure

Everything you need to runautonomous AI agents