grep#
Build typed tools from annotations and docstrings. See the guide.
Search UTF-8 text files selected by a relative or absolute glob pattern.
Parameters#
{
"additionalProperties": false,
"properties": {
"pattern": {
"minLength": 1,
"title": "Pattern",
"type": "string",
"description": "Python regular expression to search for on each line."
},
"file_pattern": {
"default": "**/*",
"minLength": 1,
"title": "File Pattern",
"type": "string",
"description": "Relative or absolute glob selecting files to search."
},
"case_sensitive": {
"default": true,
"title": "Case Sensitive",
"type": "boolean",
"description": "Whether letter case must match exactly."
},
"max_results": {
"maximum": 5000,
"minimum": 1,
"type": "integer",
"default": 200,
"description": "Maximum number of matching lines to return."
}
},
"required": [
"pattern"
],
"title": "ToolInput",
"type": "object"
}
Model-facing snippet#
grep(pattern="class\s+Agent", file_pattern="src/**/*.py", max_results=200)
Guidelines#
Use a narrow file_pattern to avoid scanning unrelated files.
Escape regular-expression characters when searching for literal text.
Each match is one path:line:text line; use read_file for surrounding context.
- zett_agent.tools.coding.grep#
A tool is a name, an input schema, and an async callable.
- 参数:
arguments (Mapping[str, Any])
- 返回类型:
Direct Python calls use await tool(arguments_mapping). The snippet
above describes model-facing invocation syntax, not the Python call signature.