increase-evals

Generate 5,000+ diverse eval test cases for boredgame.lol.

Updated Mar 25, 2026

Search Tags:#automation #quality #coverage #generate #test-cases #evals

Ready to run

npx skills add https://github.com/stopitdan/recommendagame --skill increase-evals

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: increase-evals
Source: https://github.com/stopitdan/recommendagame/tree/main/.claude/skills/increase-evals
Command: npx skills add https://github.com/stopitdan/recommendagame --skill increase-evals

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill helps expand the evaluation suite for boredgame.lol by generating additional eval test cases to reach 5,000+ total, addressing gaps in coverage and reliability.

Core Features & Use Cases

Mass-gen: automatically creates thousands of diverse eval cases using the existing generator workflow and templates.
Targeted expansion: focuses on weak categories like real-user-feedback and regression to improve robustness.
Quality and verification: provides quick checks of case counts and distribution to ensure broad coverage.

Quick Start

Run the mass-generation script to append new eval cases and verify the updated total count.

Frequently Asked Questions about increase-evals

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

FAQPage Schema

How do I generate thousands of eval test cases to improve coverage?▼

You can generate 5,000+ diverse eval test cases by running an automated mass-creation script that appends new entries to your existing eval suite. This process utilizes existing generator workflows and templates to expand test coverage across multiple categories.

What is the best way to expand an eval suite for weak categories like real-user-feedback and regression?▼

The best way to expand an eval suite for weak categories is through targeted selective hand-curated additions. This approach specifically focuses on categories like real-user-feedback and regression to improve evaluation robustness and reliability.

How do I log work and update files when adding new eval cases?▼

To log work when adding eval cases, you must update the evals/cases.json file with the new entries and record your progress in evals/EVAL-WORKLOG.md. This workflow ensures all generated test cases are tracked and integrated into the suite.

Can I verify the distribution and count of my generated eval test cases?▼

Yes, you can verify the distribution and count of your generated eval test cases using quick checks. This verification ensures broad coverage and confirms that your suite has successfully reached the 5,000+ test case target.

Does generating thousands of eval cases require any external dependencies?▼

Generating thousands of eval cases does not require external dependencies. The skill relies on internal mass-creation scripts, existing templates, and standard JSON logging workflows to achieve automated generation and verification.