smart-explore

Search code structure with tree-sitter AST parsing to locate functions and symbols.

1|Updated Jul 22, 2025
One-click install
npx skills add https://github.com/JerryGanst/qiduandebug --skill smart-explore
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: smart-explore
Source: https://github.com/JerryGanst/qiduandebug/tree/main/plugin/skills/smart-explore
Command: npx skills add https://github.com/JerryGanst/qiduandebug --skill smart-explore

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill streamlines code exploration by enabling efficient, token-optimized structural code searching, making it easier to understand code structure, locate functions, and navigate large codebases without reading entire files.

Core Features & Use Cases

  • Structural Code Search: Uses tree-sitter AST parsing to find functions, classes, and symbols within code.
  • Token Efficiency: Significantly reduces token consumption compared to reading entire files or using traditional grep/glob methods.
  • Use Case: When trying to understand how a specific feature like "user authentication" is implemented across a project, use smart_search to find all relevant functions and then smart_unfold to view the implementation of the most critical ones.

Quick Start

Use smart_search to find all symbols related to 'database connection' within the './src' directory.

Frequently Asked Questions about smart-explore

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

FAQPage Schema
How do I search for functions and classes in a codebase without reading entire files?

Structural code search uses tree-sitter AST parsing to locate functions and classes without reading entire files. It navigates the abstract syntax tree to find specific symbols, significantly reducing token consumption compared to traditional file reading methods.

What is the best way to navigate large codebases to understand how a feature is implemented?

The best way to navigate large codebases is using structural code search to find relevant symbols, followed by targeted symbol unfolding to view specific implementations. This approach reveals how features are implemented across a project while optimizing token usage.

Can I use tree-sitter AST parsing for code exploration instead of grep and glob?

Yes, tree-sitter AST parsing replaces traditional grep and glob for code exploration. It provides structural understanding of the codebase by parsing the syntax tree, enabling efficient symbol discovery and code navigation with significant token savings.

How do token-optimized code search tools help with AI code navigation?

Token-optimized code search reduces the context window size needed for AI code navigation. By parsing ASTs to extract only relevant symbols and targeted implementations instead of full files, it minimizes token consumption during code exploration and structural analysis.

Does structural code search work for finding specific symbol implementations across a project?

Yes, structural code search works for finding symbol implementations by searching the AST for functions and classes. You can then use targeted symbol unfolding to view the detailed implementation of only the most critical symbols discovered.

When should I use AST parsing for code exploration over traditional file reading?

You should use AST parsing for code exploration when navigating large codebases or when token efficiency is critical. It replaces full file reads by unfolding only targeted symbols, making it ideal for understanding complex code structures without exhausting token limits.