# Ripgrep Search

> Blazing-fast code and file search using ripgrep (rg). Recursively searches patterns across large codebases with .gitignore awareness, type filters, and context lines.

- Skill: `loonghao/ripgrep-search` (Agent Skill)
- Install (CLI): `npx skillmds@latest add loonghao/ripgrep-search`
- Raw SKILL.md: https://api.skillmd.com/api/skills/loonghao/ripgrep-search/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- License: MIT
- Author: loonghao (https://skillmd.com/u/loonghao)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/loonghao/ripgrep-search

---


# Ripgrep Search

Search code and files at blazing speed using [ripgrep](https://github.com/BurntSushi/ripgrep).

## When to use

- Finding all usages of a function or variable across a large codebase
- Locating TODO/FIXME comments across a project
- Searching within specific file types (Rust, Python, TypeScript, etc.)
- Replacing `grep -r` in CI scripts for 10-100× speed improvement

## Inputs

```json
{
  "pattern": "fn parse_",
  "path": "./src",
  "file_type": "rust",
  "context_lines": 3,
  "case_sensitive": false,
  "max_results": 50
}
```

| Field           | Required | Description                                        |
|-----------------|----------|----------------------------------------------------|
| `pattern`       | ✓        | Regex or literal pattern to search                 |
| `path`          | ✗        | Directory or file to search (default: current dir) |
| `file_type`     | ✗        | Filter by file type: `rust`, `py`, `ts`, etc.      |
| `context_lines` | ✗        | Lines of context before/after each match           |
| `case_sensitive`| ✗        | Default `false`                                    |
| `max_results`   | ✗        | Limit output lines (default: unlimited)            |

## Example commands

```bash
# Search for a pattern in Rust files
rg "fn parse_" --type rust -C 3

# Find all TODO comments
rg "TODO|FIXME|HACK" --glob "*.{rs,py,ts}"

# Search case-insensitively, limit output
rg -i "error handling" --max-count 20
```

## Output

```json
{
  "matches": [
    {
      "file": "src/skill_md.rs",
      "line": 64,
      "content": "pub fn parse_skill_md(content: &str, fallback_name: &str) -> ParsedSkillMd {"
    }
  ],
  "total_matches": 1,
  "elapsed_ms": 12
}
```