plugin

styx — Claude Code plugin: an MCP gateway to LLM APIs

Q: How do I install styx in Claude Code?

Add the marketplace first: `/plugin marketplace add anthropics/claude-plugins-community`. Then install the plugin: `/plugin install styx@claude-community`.

A plugin that runs an API gateway for several model providers and lets you drive it from Claude: routing, spend, keys.

Install into Claude Code · Plugin marketplace All methods and platforms ↓

/plugin marketplace add anthropics/claude-plugins-community
/plugin install styx@claude-community

Open on GitHub

What styx does

Styx sits between your app and the model providers. Requests go out through a single endpoint to OpenAI, Anthropic, Google or Mistral, and a failed provider gets swapped for a working one automatically. Spend shows up as it happens, and team access is split by scoped keys.

Inside Claude you get /styx:setup and /styx:status with health, usage or providers as the argument, plus the @styx-ops subagent for diagnosing trouble, trimming cost and handling keys. The MCP tools cover provider and model listings, usage and cache stats, key creation and revocation, and request routing.

Docker and Docker Compose are required, and a Styx instance has to be running already. The plugin drives that instance; it does not create one, so a missing gateway leaves the setup and status commands with nothing to talk to. Self-hosted, Apache 2.0.

What it helps with

send model requests to several providers through one endpoint
keep serving traffic when a provider goes down
watch API spend without opening a separate dashboard
hand a developer a key with limited scope
find out why the gateway is failing without leaving the editor

Example prompts for the agent

Set up and start the Styx gateway with /styx:setup
Show today's spend: /styx:status usage
@styx-ops check why requests to the provider keep failing

How to install styx

Plugin marketplace

/plugin marketplace add anthropics/claude-plugins-community

/plugin install styx@claude-community

Step by step, with commands for each platform: Claude Code plugins: marketplaces, installing, and building your own

Questions about styx

How do I install styx in Claude Code?

Add the marketplace first: /plugin marketplace add anthropics/claude-plugins-community. Then install the plugin: /plugin install styx@claude-community.

What does styx do?

It is an MCP gateway in front of several model APIs. It routes requests, tracks spending and issues scoped keys.

What does the plugin need to work?

Docker with Docker Compose and a Styx instance that is already running. The commands manage that instance rather than starting one.

Source text

claude-code-plugin · Apache-2.0

# Styx — Claude Code Plugin

The first **MCP-native AI API gateway**, packaged as a Claude Code plugin.

## Installation

```
/plugin install styx@claude-plugin-directory
```

Or: `/plugin > Discover > styx`

## Prerequisites

- Docker and Docker Compose
- A running Styx instance

## Quick Start

```
/styx:setup
```

## Commands

| Command | Description |
|---------|-------------|
| `/styx:setup` | Set up and start the Styx gateway |
| `/styx:status [health\|usage\|providers]` | Check gateway status |

## Agent

| Agent | Description |
|-------|-------------|
| `@styx-ops` | Diagnose issues, optimize costs, manage keys |

## MCP Tools

`list_providers` · `list_models` · `get_usage` · `get_health` · `create_api_key` · `revoke_api_key` · `list_api_keys` · `get_cache_stats` · `route_request`

## Links

- [Repository](https://github.com/styx-hq/styx)
- [Website](https://styxhq.com)
- License: Apache 2.0

azure-apim-ai-gateway official github A GitHub Copilot rule for building an AI gateway on Azure API Management in front of Microsoft Foundry and other LLM APIs. Useful when Copilot reaches for outdated policies and gets element order wrong. ★ 40,451repo stars llm-gateway devops-skills This skill walks you through deploying one API gateway in front of several LLMs: load balancing, rate limits, keys, caching and spend tracking. CX Codex +1 333installs jetson-llm-serve Jetson Team Creates a serving recipe for LLMs or VLMs on Jetson with vLLM or SGLang. Use it when an OpenAI-compatible model server is needed. CX Codex +1 1,200installs senior-ml-engineer Alireza Rezvani Guides production ML deployment, MLOps pipelines, RAG systems and LLM API integration. Use it when a project needs monitoring, rollout checks or request cost controls. 915installs serving-llms-on-epyc amd Deploys a vLLM endpoint with the zentorch backend on AMD EPYC CPUs using Docker, Podman, or conda. Handy for CPU-based LLM inference pinned to a single socket. CX Codex +1 819installs future-agi-platform reason-machines A skill that walks your coding agent through Future AGI: wiring up tracing, running eval metrics, and self-hosting the platform. CX Codex +1 308installs

From the same set2

styx-mcp-server This MCP server wires your client to a self-hosted Styx gateway, which spreads requests across Anthropic, Anthropic, Google and Mistral. CX Codex +1 ★ 5repo stars