ai-agent

onyx

Name: onyx
Author: onyx-dot-app

Open Source AI Platform - AI Chat with advanced features that works with every LLM

32,327 stars+67★ this week4,523 forksPythonUpdated 10/5/2026100% free · open source

What it does

Onyx is an open‑source AI chat platform that lets you run a UI and API on top of any LLM (e.g., OpenAI, Anthropic, Llama) with advanced chat features.

When to use it

•You need a self‑hosted chat UI for internal teams or customer‑facing bots.
•You want to swap between different LLM providers without rewriting code.
•You require built‑in tools like memory, file uploads, or function calling out of the box.

Ready-to-paste prompt

curl -X POST http://localhost:8000/api/chat \
  -H "Content-Type: application/json" \
  -d '{"message":"Draft a 2‑sentence pitch for a fintech startup that uses AI to detect fraud in real time."}'

Heads up: Onyx requires a valid LLM API key (e.g., OpenAI, Anthropic) set in the `.env` file; without it the server will start but every chat request returns a 401 error.

Saves to your device

Use with Claude

New

Skip the builder — one click puts this in Claude, Cursor, Antigravity and more.

🛠️ Technical setup

Expect 20–40 minutes in a terminal — or let your AI agent drive it.

Try it instantly — no install

Claude Code

mkdir -p ~/.claude/skills/0dd34580-2886-4142-b0e3-71f7240e6671 && curl -fsSL https://workflowstacks.com/api/skills/0dd34580-2886-4142-b0e3-71f7240e6671/claude-skill -o ~/.claude/skills/0dd34580-2886-4142-b0e3-71f7240e6671/SKILL.md

Open in another AI app

Opens the app with this repo with the prompt ready to go — no copy-paste needed.

Connect the whole catalog (MCP)

claude mcp add --transport http workflowstacks https://workflowstacks.com/api/mcp

Adds a WorkflowStacks connector to Claude Code: search and load any skill here by chatting.

How onyx works

Codeflow

Free to inspect

Onyx is a very large Python project (~720k lines across 4880 code files, plus 1782 test files). It is a full software project: use it through its install path rather than reading it end to end. Last commit this month, Other license, has a test suite.

Size

Very large codebase

~720k lines · 4880 code files · days to read — use, don't read

Setup

Developer setup

A real software project. Use it via its install path; don't expect to read it all.

Runs on

Python · Node.js

No API keys detected

Python 64%TypeScript 26%Go 6%JavaScript 1%HCL 1%

Where to start reading

1
README.md
Start here — what it does and how to install it
2
AGENTS.md
The instructions the AI actually follows
3
cli/main.go
Where the program starts running
4
package.json
Dependencies and the commands it exposes
5
examples/assistants-api/topics_analyzer.py
A worked example — copy this to get going

What's in each folder

backend/Backend / API3689 files

web/Frontend / UI2458 files

cli/Command-line entry points170 files

extensions/Sub-packages / plugins27 files

integrations/Sub-packages / plugins7 files

docs/Documentation76 files

examples/Examples you can copy13 files

mobile/Folder452 files

READMEHas testsDocumentedCI checksExamples includedOther licenseUpdated this month

Use with Claude

New

Skip the builder — one click puts this in Claude, Cursor, Antigravity and more.

🛠️ Technical setup

Expect 20–40 minutes in a terminal — or let your AI agent drive it.

Try it instantly — no install

Claude Code

mkdir -p ~/.claude/skills/0dd34580-2886-4142-b0e3-71f7240e6671 && curl -fsSL https://workflowstacks.com/api/skills/0dd34580-2886-4142-b0e3-71f7240e6671/claude-skill -o ~/.claude/skills/0dd34580-2886-4142-b0e3-71f7240e6671/SKILL.md

Open in another AI app

Opens the app with this repo with the prompt ready to go — no copy-paste needed.

Connect the whole catalog (MCP)

claude mcp add --transport http workflowstacks https://workflowstacks.com/api/mcp

Adds a WorkflowStacks connector to Claude Code: search and load any skill here by chatting.

Quick Actions

Details

Creator

onyx-dot-app

Language

Python