Start

Quickstart

Send a skill-aware request through an OpenAI-compatible client.

Before you start

Create an account and add prepaid credits.
Create a test API key and copy it when it is shown.
Store the key as SKILLROUTER_API_KEY in your server environment.
Install the OpenAI SDK for your language.

A complete SkillRouter key is shown only after create or rotate. Never put it in browser code or commit it to source control.

A sr_test_* key separates development credentials and spend limits, but it still sends real provider requests and consumes prepaid credits. A sr_live_* key always uses production routing and commercial gates, even if the API server environment is configured incorrectly.

Make the first request

import os

from openai import OpenAI

client = OpenAI(
    api_key=os.environ["SKILLROUTER_API_KEY"],
    base_url="https://api.useskillrouter.com/v1"
)

response = client.chat.completions.create(
    model="anthropic/claude-sonnet",
    messages=[
        {"role": "user", "content": "Audit this page for SEO."}
    ],
    extra_body={"skill": "seo-audit"}
)

print(response.choices[0].message.content)

The user message is preserved. SkillRouter resolves the published skill version, checks model capabilities, compiles higher-priority instructions, selects a provider route, and records metered usage.

Use a virtual model

Replace the model with skillrouter/seo-audit and omit the skill field. The virtual-model registry selects the skill version, preferred model, and permitted fallback policy.

{
  "model": "skillrouter/seo-audit",
  "messages": [{"role": "user", "content": "Audit example.com"}]
}