Open Source AI Agent Framework
OnyxAgent

Your Super AI Assistant

OnyxAgent proactively plans tasks, controls your computer and external services, creates and runs Skills, builds a personal knowledge base and long-term memory, and grows alongside you through self-evolution.

Terminal
$git clone https://github.com/AkshayCoder48/OnyxAgent.git && cd OnyxAgent && pip install -r requirements.txt && python app.py
Capabilities

Everything You Need in an AI Agent

OnyxAgent is a complete Agent Harness — every layer is decoupled and independently extensible.

Planning & Reasoning

Decomposes complex tasks and executes them step by step, looping over tools until the goal is reached.

Three-Tier Memory

Context → Daily → Core architecture with automatic Deep Dream distillation and hybrid keyword + vector retrieval.

Knowledge Base

Auto-curates structured knowledge into a Markdown wiki, builds an evolving knowledge graph with visual browsing.

Self-Evolution

Reviews conversations automatically to improve skills, follow up on unfinished tasks, and consolidate memory.

Skills System

One-click install from Skill Hub, GitHub; or create custom skills via natural-language conversation.

Built-in Tools

File I/O, terminal, browser, scheduler, memory retrieval, web search, and 10+ more tools with MCP integration.

Multi-Channel

Web, WeChat, Feishu, DingTalk, WeCom, QQ, Telegram, Slack, Discord and more.

Multimodal

First-class support for text, images, voice, and files — recognition, generation, and delivery.

Multi-Model

Claude, GPT, Gemini, DeepSeek, Qwen, GLM, Kimi, MiniMax, Doubao — swap providers from the Web console.

Get Started

Up and Running in Minutes

A one-line installer takes care of dependencies, configuration, and startup.

1Clone the repository

bashgit clone https://github.com/AkshayCoder48/OnyxAgent.git
cd OnyxAgent
pip install -r requirements.txt
python app.py

2Open the Web Console

bash# Open http://localhost:9899 in your browser

3Start chatting!

The Web console is your one-stop hub to chat with the Agent, configure models, connect channels, and install skills.

Deploying on a server?

Set web_host to 0.0.0.0 in config.json to make the console reachable from outside, and set web_password to protect it. Don't forget to open port 9899 in your firewall.

Command Line

Onyx CLI Reference

Manage your OnyxAgent service with the powerful Onyx CLI.

onyx start

Start the OnyxAgent service

Service
onyx stop

Stop the OnyxAgent service

Service
onyx restart

Restart the OnyxAgent service

Service
onyx status

Show service status

Service
onyx logs

View service logs

Service
onyx update

Pull latest code and restart

Service
onyx skill install <name>

Install a skill from the hub

Skills
onyx skill list

List installed skills

Skills
onyx skill create <name>

Create a new skill

Skills
onyx install-browser

Install browser automation

Setup
onyx context

Manage conversation context

Memory
onyx knowledge

Manage knowledge base

Memory

Install the Onyx CLI

bash# After installing OnyxAgent, the onyx CLI is available globally
onyx --help
onyx --version
Architecture

Complete Agent Harness

Every layer is decoupled and independently extensible. Messages flow through Channels, the Agent Core plans and reasons, Models generate responses.

Channels

Messages flow in through Channels

Web ConsoleWeChatFeishuDingTalkTelegramSlackDiscord

Agent Core

Plans and reasons over memory, knowledge, and tools

PlanningMemory (3-tier)KnowledgeSkillsSelf-Evolution

Models

Generate the response via any LLM provider

ClaudeGPTGeminiDeepSeekQwenGLMMiniMax

Tools

Built-in tools with MCP integration

File I/OTerminalBrowserWeb SearchSchedulerMCP
Models

Bring Your Own Model

Swap LLM providers from the Web console with one click. Support for all major providers.

OpenAI

GPT-4o

GPT-4 Turbo

GPT-3.5

Claude

Claude 3.5 Sonnet

Claude 3 Opus

Claude 3 Haiku

Gemini

Gemini Pro

Gemini Ultra

DeepSeek

DeepSeek-V3

DeepSeek-R1

Qwen

Qwen-Max

Qwen-Plus

Qwen-Turbo

GLM

GLM-4

GLM-4V

Kimi

Moonshot-v1

MiniMax

abab6.5

Doubao

Doubao-pro

Doubao-lite

Custom

Any OpenAI-compatible provider

Channels

Connect Everywhere

Integrate with all major IM platforms and web interfaces.

Web Console

Built-in web UI at localhost:9899

WeChat

Personal & official accounts

Feishu

Lark/Feishu bot integration

DingTalk

DingTalk robot integration

WeCom

Enterprise WeChat

QQ

QQ channel integration

Telegram

Telegram bot support

Slack

Slack app integration

Discord

Discord bot support

WeChat MP

WeChat Official Account

WeChat KF

WeChat Customer Service

Terminal

Command-line interface

Tools

Built-in Tool Arsenal

14+ built-in tools with native MCP integration for unlimited extensibility.

bash

Execute shell commands

read

Read file contents

write

Write to files

edit

Edit existing files

ls

List directory contents

browser

Web browser automation

web-search

Search the web

web-fetch

Fetch web page content

memory

Retrieve from memory

scheduler

Schedule tasks

vision

Analyze images

send

Send messages/files

env-config

Manage environment config

mcp

Model Context Protocol

Deployment

Deploy Anywhere

Multiple deployment modes — local, Docker, or cloud server.

Local
Run on your personal machine
bash# Clone the repository
git clone https://github.com/AkshayCoder48/OnyxAgent.git
cd OnyxAgent

# Install dependencies
pip install -r requirements.txt

# Start OnyxAgent
python app.py
Docker
Containerized deployment
bash# Clone & build from source
git clone https://github.com/AkshayCoder48/OnyxAgent.git
cd OnyxAgent
docker build -t onyxagent .
docker run -p 9899:9899 onyxagent
Cloud Server
Production deployment
bash# Set web_host for external access
# In config.json:
{
  "web_host": "0.0.0.0",
  "web_password": "your-password"
}

# Open port 9899 in firewall
# Access from anywhere
FAQ

Frequently Asked Questions