explore_skill
- Python
6
GitHub Stars
1
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill bahayonghang/my-claude-code-settings --skill explore- SKILL.md5.5 KB
Overview
This skill is a fast, precise codebase search agent that locates files, functions, and patterns inside the current project. It helps you find where features live, trace call chains, and understand module structure so you can act quickly on code tasks. The agent returns actionable paths, snippets, and suggested next steps.
How this skill works
It runs parallel searches using filename globs, content grep, LSP symbol/definition lookups, AST pattern matching, and Git history queries to triangulate results. The search strategy narrows first (exact matches) then broadens, preferring structure (directories, symbols) before raw text. Results include file paths, line numbers, short code excerpts, and provenance (definition vs. reference vs. commit).
When to use it
- Find where a specific feature or handler is implemented (e.g., authentication, payments).
- Search for all occurrences of a function, class, variable, or pattern (e.g., console.log, TODOs).
- Trace a function’s definition and all call sites to understand dependencies and impact.
- Map module structure and identify which files form a subsystem.
- Investigate who introduced a change or when a line was added using Git history.
Best practices
- Start with precise names or paths; relax patterns if no hits (narrow→wide).
- Combine symbol/LSP queries with content grep for both definitions and loose matches.
- Exclude noisy dirs (node_modules, .git) to speed searches and reduce false positives.
- When tracing calls, locate the definition first then find references to avoid missing indirect uses.
- Request file excerpts by path:line to get minimal, runnable context quickly.
Example use cases
- "Where is user authentication implemented?" → returns auth directory, files, and main functions with snippets.
- "Find all uses of handleSubmit" → shows definition location and all call sites with context lines.
- "Locate every console.log or debug statement" → AST or grep results grouped by file and line.
- "Which file added this code and when?" → git log/blame output for the target lines and commits.
- "Show the module graph for payment-related files" → directory tree and key imports/exports list.
FAQ
You receive a concise file list, file:line code snippets, and notes indicating definition vs. reference and Git provenance when requested.
How fast and comprehensive are searches?
Searches run in parallel across multiple techniques for speed and coverage; initial results are returned quickly and can be refined iteratively.