Grep

Search file contents using ripgrep.

Package: @agentrail/capabilities

Parameters

Name	Type	Required	Description
pattern	string	Yes	Regular expression pattern to search for.
path	string	No	File or directory to search in. Defaults to the current working directory.
glob	string	No	Glob pattern to filter files (e.g. "*.ts", "**/*.{ts,tsx}").
type	string	No	File type to search (e.g. "js", "py", "rust"). More efficient than glob for standard file types.
output_mode	"content" | "files_with_matches" | "count"	No	Output format. Defaults to "files_with_matches".
multiline	boolean	No	Enable multiline mode where patterns can span lines. Default: false.
case_insensitive	boolean	No	Case-insensitive search. Default: false.

Output modes

Value	Description
"files_with_matches"	Return only the paths of matching files (default).
"content"	Return matching lines with file and line number context.
"count"	Return the number of matches per file.

Result

{ matches: string[] }

Returns "(no matches)" when nothing is found.

Usage notes

• Always use Grep for code search. Never call grep or rg via Bash.
• Literal braces require escaping in regex: use interface\{\} to find interface{}.
• For cross-line patterns use multiline: true with a pattern like struct \{[\s\S]*?field.
• Combine type and pattern for efficient language-specific searches.

Example

// Find all TypeScript files exporting a function named "createTool"
{ pattern: "export function createTool", type: "ts", output_mode: "files_with_matches" }

// Show matching lines with context
{ pattern: "TODO|FIXME", path: "/workspace/src", output_mode: "content" }

// Case-insensitive search across markdown files
{ pattern: "permission", glob: "**/*.md", case_insensitive: true, output_mode: "content" }

Sandbox variant

The sandboxed Grep tool runs rg inside the container. Usage is identical; path is relative to /workspace by default.

Related

• Glob
• Bash