Documentation
Everything you need to integrate SkillRouter into your AI agents. From quick starts to advanced patterns.
Getting Started
Quick Start Guide
Get up and running with SkillRouter in under 5 minutes. Learn the basics of skill discovery and execution.
Authentication
Learn how to authenticate your requests using API keys and manage access tokens securely.
Your First Skill Call
Make your first skill execution and understand the request/response format.
Core Concepts
Skills & Tools
Understand the difference between skills and tools, and how they're organized in the registry.
Skill Discovery
Learn how vector-based skill discovery works and how to find the right skill for any task.
Execution Proxy
How SkillRouter handles execution, retries, authentication, and browser instances.
Skillsets & MCPs
Combine multiple skills and MCPs into reusable skillsets for complex workflows.
SDKs & API Reference
Python SDK
Official Python SDK with full type hints, async support, and comprehensive documentation.
TypeScript SDK
First-class TypeScript support with full type safety and modern ES modules.
REST API Reference
Complete API reference with examples for all endpoints and error codes.
from skillrouter import SkillRouter
# Initialize the client
sr = SkillRouter(api_key="sk_...")
# Discover the best skill for a task
skill = sr.discover("send an email to user@example.com")
# Execute the skill
result = sr.execute(
skill_id=skill.id,
params={
"to": "user@example.com",
"subject": "Hello from SkillRouter",
"body": "This email was sent by an AI agent!"
}
)
print(f"Execution successful: {result.success}")
print(f"Response: {result.data}")Need help?
Can't find what you're looking for? Our team is here to help.