shinka-setup

Generate ShinkaEvolve task scaffolds with evaluate.py and initial templates.

1|1|Updated Mar 5, 2026
One-click install
npx skills add https://github.com/tan-yong-sheng/GrowChat --skill shinka-setup-tan-yong-sheng
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: shinka-setup
Source: https://github.com/tan-yong-sheng/GrowChat/tree/main/.claude/skills/shinka-setup
Command: npx skills add https://github.com/tan-yong-sheng/GrowChat --skill shinka-setup-tan-yong-sheng

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires shinka, pyyaml, and includes scripts (resource) components.

What problem does it solve?

Generate ready-to-run ShinkaEvolve task scaffolds from a user-provided task description, including evaluate.py and initial.<ext> templates.

Core Features & Use Cases

  • Generates multi-language initial program templates and a deterministic evaluation harness for code-evolution tasks.
  • Produces stable input/output contracts and evolve-ready placeholders that can be customized by the user.
  • Suitable for rapid prototyping of optimization problems and reproducible ShinkaEvolve experiments.

Quick Start

Provide a task description and a target language, and the skill will create evaluate.py and initial.<ext> templates.

Frequently Asked Questions about shinka-setup

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

FAQPage Schema
How do I generate an evaluation harness for code evolution tasks?▼

Multi-language code evolution tasks require deterministic evaluation harnesses to measure program fitness consistently. This skill generates stable input/output contracts and evolve-ready templates from a task description to ensure reproducible optimization experiments.

How do I scaffold a ShinkaEvolve task for rapid prototyping?▼

To scaffold a ShinkaEvolve task, provide a clear task description and your chosen programming language. The tool automatically produces an evaluate.py file and a multi-language initial program template with customizable placeholders.

What do I need to generate multi-language initial program templates with this tool?▼

You need a clear task description and a chosen target language to generate multi-language initial program templates. Python dependencies including the shinka package and pyyaml must also be installed in your environment.

Can I customize the initial program templates after they are generated?▼

Yes, the generated initial.<ext> templates include stable input/output contracts and evolve-ready placeholders. You can customize these placeholders to fit your specific optimization problem before running your ShinkaEvolve experiments.

Why does my generated evaluate.py fail to run?▼

Generated evaluate.py files require the shinka and pyyaml dependencies to function properly. Ensure both packages are installed in your Python environment and that your initial.<ext> template matches the expected input/output contracts.