use-understudy-gateway

Route developer workloads through the Understudy gateway or manage local keys.

10|5|Updated Jun 3, 2026
One-click install
npx skills add https://github.com/understudylabs/understudy-agent-tools --skill use-understudy-gateway
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: use-understudy-gateway
Source: https://github.com/understudylabs/understudy-agent-tools/tree/main/skills/use-understudy-gateway
Command: npx skills add https://github.com/understudylabs/understudy-agent-tools --skill use-understudy-gateway

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Simplifies running inference and evaluation workloads by routing through the Understudy gateway or by managing local provider keys, giving developers a choice between gateway-backed execution and local-only runs.

Core Features & Use Cases

  • Handles login, project/key management, model routing, and durable runs via the CLI.
  • Enables seamless switching between local provider keys and the gateway route for flexible workflow control.
  • Guides gateway-based inference, A/B routing, capture, and deployment planning in development and testing environments.

Quick Start

Run the understudy CLI to login, configure projects/keys, and select gateway routing or local keys for your workload.

Frequently Asked Questions about use-understudy-gateway

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I route inference workloads through a gateway from the CLI?

You can route inference workloads through a gateway by running the understudy CLI to login, configure projects and keys, and select gateway routing for durable command execution. It enforces an always-stream rule for gateway inferences.

Can I use local provider keys instead of gateway routing for evaluation runs?

Yes, you can use local provider keys instead of gateway routing. The CLI enables seamless switching between local-only execution and gateway-backed routing to give you flexible workflow control.

What is gateway-based model routing and how does it handle credentials?

Gateway-based model routing directs developer workloads through the Understudy gateway while applying safety gates around credential handling. It manages local keys and guides A/B routing, capture, and deployment planning.

Does the gateway inference path support trace capture and deployment workflows?

Yes, the gateway inference path supports trace capture and deployment workflows. It guides trace capture during durable command execution and assists with deployment planning in development and testing environments.

What is the best way to manage projects and keys for CLI workload execution?

The best way to manage projects and keys is using the understudy CLI. It handles login, project and key management, and model routing to securely execute durable commands across local and gateway paths.

When should I not use the gateway route for my inference tasks?

You should avoid the gateway route when you prefer local-only execution. The system applies an always-stream rule for gateway inferences and enforces safety gates, so local keys offer a simpler alternative for flexible workflow control.