smart-explore

Locate symbols and outline file structure using tree-sitter AST queries.

Updated Apr 12, 2026

Search Tags:#tree-sitter #ast #token-efficiency #symbol-search #code-exploration #code-outline

Ready to run

npx skills add https://github.com/trezero/archon-trinity --skill smart-explore-trezero

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: smart-explore
Source: https://github.com/trezero/archon-trinity/tree/main/integrations/claude-code/plugins/archon-memory/skills/smart-explore
Command: npx skills add https://github.com/trezero/archon-trinity --skill smart-explore-trezero

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill provides token-efficient, AST-powered exploration to avoid reading entire files when locating functions, understanding structure, or inspecting unfamiliar codebases.

Core Features & Use Cases

AST-powered Search: Finds ranked symbol matches with signatures and file locations to quickly locate where behavior is implemented.
Folded Outlines: Returns file outlines with function, class, and method signatures while folding bodies to summarize structure.
Targeted Unfolding: Extracts the full source for a single symbol so you can read only the relevant code; for example, locating and inspecting an authentication function across a large repository.

Quick Start

Ask smart-explore to perform a smart_search for "authenticate user" scoped to src/ to return ranked symbol matches and file locations.

Frequently Asked Questions about smart-explore

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

FAQPage Schema

How do I explore a large codebase to find functions without reading entire files?▼

AST-powered code exploration locates ranked symbol matches with signatures and file locations, allowing you to inspect file structure and find functions without reading full files.

Can I get a code outline for Python and TypeScript files using AST queries?▼

Yes, AST-based queries generate folded outlines returning function, class, and method signatures while folding bodies to summarize structure for languages like Python, JavaScript, TypeScript, Go, Rust, Java, Ruby, C, and C++.

What is the best way to minimize token usage when searching for symbols in an unfamiliar repository?▼

Token-efficient AST code exploration minimizes token usage by returning ranked symbol search results, folded outlines, and per-symbol source extracts instead of reading entire files.

How do I extract the full source code of a single function from a large repository?▼

Targeted unfolding extracts the full source for a single symbol, allowing you to read only the relevant code by locating and inspecting specific functions across large repositories.

Does tree-sitter AST code exploration work with Go, Rust, and C++?▼

Yes, tree-sitter AST code exploration supports Go, Rust, C, and C++, alongside Python, JavaScript, TypeScript, Java, and Ruby, for symbol search and file outlining.

Why should I use AST-based code exploration instead of reading full source files?▼

AST-based code exploration provides token-efficient codebase navigation by returning ranked symbol matches and folded outlines, preventing the token waste associated with reading entire files in large or unfamiliar repositories.