chore: update quickstart to use new llmjudge format

[Cali0707]

Summary by CodeRabbit

• Refactor
  • Restructured agent configuration management to use file-based references for improved organization and maintainability
  • Updated evaluation configurations to load agent and judge settings from external files instead of environment variables
  • Applied configuration improvements across multiple example iterations demonstrating best practices

[coderabbitai]

Caution

Review failed

The pull request is closed.

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro

Run ID: 842ccc63-3d08-4883-b72d-05187e58e6a4

📥 Commits

Reviewing files that changed from the base of the PR and between 2925f22 and 065dbd5.

📒 Files selected for processing (4)

• 03-evolution-case-study/iteration-1-bad/evals/agent.yaml
• 03-evolution-case-study/iteration-1-bad/evals/eval.yaml
• 03-evolution-case-study/iteration-2-good/evals/agent.yaml
• 03-evolution-case-study/iteration-2-good/evals/eval.yaml

---

📝 Walkthrough

Walkthrough

The changes introduce agent configuration files and update evaluation manifests across two iteration directories, transitioning from environment-variable-based agent and judge configuration to externalized file-based YAML references.

Changes

Cohort / File(s)	Summary
Iteration 1 (Bad) - Agent Setup 03-evolution-case-study/iteration-1-bad/evals/agent.yaml	New Agent resource manifest defining claude-code-acp agent with acp.cmd: "claude-agent-acp".
Iteration 1 (Bad) - Evaluation Config 03-evolution-case-study/iteration-1-bad/evals/eval.yaml	Modified eval configuration to reference agent.yaml file for agent setup and llmJudge config, replacing environment-variable-based settings.
Iteration 2 (Good) - Agent Setup 03-evolution-case-study/iteration-2-good/evals/agent.yaml	New Agent resource manifest defining claude-code-acp agent with acp.cmd: "claude-agent-acp".
Iteration 2 (Good) - Evaluation Config 03-evolution-case-study/iteration-2-good/evals/eval.yaml	Modified eval configuration to reference agent.yaml file for agent setup and llmJudge config, replacing environment-variable-based settings.

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~10 minutes

Poem

🐰 From env vars scattered wide and far,
Configuration hops to files—a star!
Two iterations show the dance of change,
From chaotic env to manifest range.
Better organized, cleaner, more sane,
The rabbit approves this refactored lane! ✨

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}