test_props

Automate end-to-end props testing workflows with Podman and host networking.

4|1|Updated May 19, 2025
One-click install
npx skills add https://github.com/agentydragon/ducktape --skill test-props
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: test_props
Source: https://github.com/agentydragon/ducktape/tree/main/.claude/skills/test_props
Command: npx skills add https://github.com/agentydragon/ducktape --skill test-props

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires openssl, psql, podman, curl.

What problem does it solve?

Run end-to-end props testing to validate critic, grader, improver, and optimizer workflows in a Podman + host-networked environment using a real API key.

Core Features & Use Cases

  • Orchestrates infrastructure setup, DB initialization, backend startup, and agent image pushes to validate end-to-end prop testing workflows.
  • Verifies critic, grader, and improver components behave correctly across a full system run.
  • Use Case: Simulate a real evaluation session by running the full pipeline in a local environment to ensure all pieces integrate.

Quick Start

Execute the test_props workflow with the all target to run the complete end-to-end test suite.

Frequently Asked Questions about test_props

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

FAQPage Schema
How do I run end-to-end testing for critic, grader, and improver workflows with an OpenAI API key?

End-to-end testing for critic, grader, and improver workflows is automated by orchestrating infrastructure setup, database initialization, backend startup, and agent image pushes using a real OpenAI API key in a Podman environment.

What prerequisites do I need for Podman-based end-to-end props testing?

Podman-based end-to-end props testing requires Podman, a local OCI registry, openssl, psql, curl, and a configured OPENAI_API_KEY. Configuration is specified in a .claude/skills/test_props/config.podman.toml file for host-networked environments.

How does infrastructure automation work for local backend testing with Podman?

Infrastructure automation for local backend testing coordinates Podman host-networking setup, database initialization, backend startup, and agent image pushes to a local OCI registry to exercise the full evaluation pipeline.

Can I validate optimizer workflows in a local host-networked environment?

Yes, optimizer workflows can be validated in a local host-networked environment by running the full end-to-end test suite, which simulates a real evaluation session to ensure all components integrate correctly.

What is the best way to test an agent evaluation pipeline locally before deployment?

The best way to test an agent evaluation pipeline locally is running the complete end-to-end test suite with the all target, which validates critic, grader, improver, and optimizer workflows using a real API key and local OCI registry.

Why does Podman end-to-end testing require a local OCI registry?

Podman end-to-end testing requires a local OCI registry to facilitate agent image pushes, which are necessary to exercise the critic, grader, improver, and optimizer workflows across the full system run.