ck

Searches codebases by meaning and keywords with JSONL output for AI agents.

1|Updated Jan 14, 2026
One-click install
npx skills add https://github.com/r0227n/agent-mobile --skill ck
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ck
Source: https://github.com/r0227n/agent-mobile/tree/main/.claude/skills/ck
Command: npx skills add https://github.com/r0227n/agent-mobile --skill ck

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill replaces traditional command-line tools like grep and rg with a more powerful and semantically aware code search engine, ensuring more accurate and relevant code discovery.

Core Features & Use Cases

  • Semantic Code Search: Find code based on meaning, not just keywords.
  • Hybrid Search: Combine semantic understanding with keyword matching for precise results.
  • Tool Override: Enforces the use of ck over grep/rg for all code search tasks.
  • Use Case: Quickly locate all instances of error handling logic across a large codebase using natural language queries.

Quick Start

Use the ck skill to find the implementation of the 'XCUITestClient' struct in the current directory.

Frequently Asked Questions about ck

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

FAQPage Schema
How do I perform semantic code search using natural language queries?

Semantic code search allows you to find code implementations and conceptual patterns based on meaning rather than exact keywords. You can use natural language queries to locate function definitions and error handling logic across your codebase.

What is the best way to search for code implementations without knowing exact keywords?

Hybrid code search combines semantic understanding with keyword matching to find code implementations. This approach ensures more accurate code discovery than traditional grep tools by interpreting the intent behind your natural language queries.

Can I use semantic search output for programmatic consumption in AI agents?

Yes, semantic code search supports JSONL output specifically designed for programmatic consumption by AI agents. This allows automated tools to process search results, filter them, and sort them for code navigation tasks.

How does semantic code search compare to traditional grep or rg tools?

Semantic code search replaces traditional grep and rg tools by offering semantic and hybrid search capabilities. Unlike grep, it finds code based on meaning and conceptual patterns using natural language, yielding more relevant results.

Does semantic code search enforce its use over grep for all search tasks?

Yes, the tool override feature enforces the use of semantic search over traditional grep or rg for all code search tasks. This ensures AI agents consistently leverage semantic understanding for code navigation and discovery.

What are the limitations of using semantic search for code navigation?

Semantic code search is designed for AI agents to find implementations and patterns within a codebase using natural language. While it offers filtering and sorting options, its effectiveness depends on the clarity of the semantic queries provided.