skills.rest
2233admin
2233adminCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

shinka-inspect

Extract top-ranked Shinka programs from a run database into a Markdown context bundle.

Updated Apr 10, 2026
Search Tags:#markdown#context#ranking#inspect#programs#shinka
One-click install
npx skills add https://github.com/2233admin/ShinkaEvolve-fork --skill shinka-inspect-2233admin
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: shinka-inspect
Source: https://github.com/2233admin/ShinkaEvolve-fork/tree/main/skills/shinka-inspect
Command: npx skills add https://github.com/2233admin/ShinkaEvolve-fork --skill shinka-inspect-2233admin

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

Analyzing and selecting top-performing Shinka programs from a completed run can be time-consuming and error-prone; this skill automates extracting the strongest programs and packaging them into a ready-to-use context bundle for iterative agent planning.

Core Features & Use Cases

  • Rank programs by combined_score and select the top-k, prioritizing correct results when available to ensure high-quality context for next steps.
  • Generate a single Markdown bundle that includes run metadata, a ranking table, per-program details, and optional feedback to guide mutation planning.
  • Produce an agent-ready context artifact that can be loaded directly by coding agents to seed subsequent tasks.

Quick Start

Run this skill on a completed Shinka run to generate a compact context bundle for agent mutation planning.

Frequently Asked Questions about shinka-inspect

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

FAQPage Schema
How do I extract top-performing programs from a SQLite database for agent context?▼

To extract top-performing programs from a SQLite database, this skill ranks records by combined_score, prioritizes correct results, and truncates code to generate a concise Markdown context bundle for agents.

What is the best way to rank Shinka programs by combined score for mutation planning?▼

Ranking Shinka programs by combined score is handled automatically by selecting the top-k entries from the run database, preferring correct programs, and packaging their snippets and metadata into a Markdown artifact.

How do I package run metadata and program snippets into a Markdown artifact?▼

Packaging run metadata and program snippets into a Markdown artifact involves extracting top-k programs from the SQLite database and compiling their details, ranking tables, and optional feedback into a single context file.

Do I need pandas to load Shinka program results into an agent context?▼

Yes, you need pandas installed as a dependency to process the run database, rank the programs, and load the resulting top-k Shinka program context bundle into your coding agent.

Can I include feedback and program details when generating a context bundle for coding agents?▼

Yes, you can include feedback and program details when generating a context bundle, as the skill optionally appends this metadata to the Markdown artifact to guide subsequent agent mutation planning.

Why does my Shinka program context bundle truncate code snippets?▼

Code snippets in the Shinka program context bundle are truncated to fit context window limitations, ensuring the generated Markdown artifact remains concise and readable for the coding agent.

Related Skills

Explore More Software Engineering Skills→
nhatnguyen1122nhatnguyen1122

shinka-inspect

Extract top Shinka programs into a compact Markdown bundle.

Community
Intermediate
adiomasadiomas

initializer-protocol

Initialize autonomous coding tasks with project detection and state machine setup.

Community
Basic
davpatel605-beepdavpatel605-beep

darwinian-evolver

Optimize prompts, regex patterns, SQL queries, and code snippets through evolutionary search loops.

Community
Advanced
tan-yong-shengtan-yong-sheng
1

shinka-inspect

Rank Shinka programs by combined_score and emit a Markdown context bundle.

Community
Intermediate