ripgrep_skill

Overview

This skill provides fast, line-oriented text search across files and directories using ripgrep (rg). It is optimized for codebases, large documents, and situations where grep or find are too slow. The skill prioritizes speed, respects .gitignore by default, and exposes common rg options for targeted searches.

How this skill works

The skill runs ripgrep queries with selectable flags for regex, fixed strings, file types, globs, and output formats. It can filter by file size, depth, hidden files, and follow symlinks. Results can be returned as matching lines, filenames, counts, JSON, or vim-compatible output for downstream tools.

When to use it

• Search for occurrences of a function, variable, or string across a codebase
• Scan large logs, books, or documents for specific lines without reading full files
• Find files that contain or lack a pattern
• Extract matching tokens or count matches for analysis
• Replace-preview of simple text occurrences across many files

Best practices

• Be specific with search paths or use --max-depth to limit scope
• Use -t filetype or --type-list instead of broad globs when possible
• Prefer -F for fixed-string searches to gain performance when regex is unnecessary
• Respect .gitignore; only use --no-ignore when you deliberately need excluded files
• Limit output with -m or -l to avoid overwhelming downstream tools

Example use cases

• Find all definitions of a function in a repository: rg "def \w+(" -t py
• Locate TODOs across a project: rg "(TODO|FIXME):"
• Search a large log for the first 10 error occurrences: rg -m 10 "ERROR" huge.log
• List files missing a license header: rg --files-without-match "Copyright" src/
• Extract all unique uppercase acronyms from documents: rg -o "\b[A-Z]{2,}\b" | sort -u