eval

Launch Loopyard projects, monitor setup agents, and record results to evals runs.

21|2|Updated Mar 15, 2026
One-click install
npx skills add https://github.com/loopyard/loopyard --skill eval-loopyard
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: eval
Source: https://github.com/loopyard/loopyard/tree/main/.claude/skills/eval
Command: npx skills add https://github.com/loopyard/loopyard --skill eval-loopyard

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Automates end-to-end evaluations by launching Loopyard projects, monitoring the setup agent, and recording results to evals runs.

Core Features & Use Cases

  • Orchestrates Docker-based project setup: seeds volumes, builds containers, and runs the agent to configure the environment.
  • Monitors progress and collects structured results (outcome, duration, tool calls, errors, service status) to the evals/runs directory.
  • Supports asynchronous execution and provides introspection commands for live state during an eval.

Quick Start

Jack into the running server and trigger an eval against a designated project path.

Frequently Asked Questions about eval

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

FAQPage Schema
How do I automate end-to-end project setup evaluations for Docker environments?

End-to-end evaluations are automated by launching Loopyard projects, building containers, and monitoring the setup agent to record structured results to evals runs without human intervention.

What programming stacks are supported for Docker-based end-to-end evaluations?

Docker-based end-to-end evaluations support Rails, Python, Node, and Go stacks by seeding volumes, building containers, and running the agent to configure the environment.

How do I monitor progress and collect results during an asynchronous eval run?

During an asynchronous eval run, you can use introspection commands to monitor live state, while the system collects structured results like outcome, duration, tool calls, and service status to the evals/runs directory.

Do I need a running Loopyard server to execute automated project evaluations?

Yes, automated project evaluations require a running Loopyard server and access to evals/project paths to successfully trigger and record asynchronous test runs.

What structured results are recorded when running end-to-end DevOps evaluations?

End-to-end DevOps evaluations record structured results including outcome, duration, tool calls, errors, and service status directly to the evals/your-project/runs directory.

Can I run zero-human-intervention evaluations across multiple project paths?

Yes, you can trigger asynchronous evaluations against designated project paths, allowing the setup agent to autonomously configure environments and record results without human intervention.