Skip to content

Open-source Python library for AI browser agents, plus Browser Use Cloud: stealth browsers and a hosted agent API.

Merge branch 'main' into schema-enforced-extraction

Saurav Pandaba68695d654a
diff --cc browser_use/tools/views.py
index 654fcd656,4eb7edb4f..e87ed840b
--- a/browser_use/tools/views.py
+++ b/browser_use/tools/views.py
@@@ -13,12 -12,27 +13,31 @@@ class ExtractAction(BaseModel)
start_from_char: int = Field(
default=0, description='Use this for long markdowns to start from a specific character (not index in browser_state)'
)
+ output_schema: SkipJsonSchema[dict | None] = Field(
+ default=None,
+ description='Optional JSON Schema dict. When provided, extraction returns validated JSON matching this schema instead of free-text.',
+ )
+ class SearchPageAction(BaseModel):
+ pattern: str = Field(description='Text or regex pattern to search for in page content')
+ regex: bool = Field(default=False, description='Treat pattern as regex (default: literal text match)')
+ case_sensitive: bool = Field(default=False, description='Case-sensitive search (default: case-insensitive)')
+ context_chars: int = Field(default=150, description='Characters of surrounding context per match')
+ css_scope: str | None = Field(default=None, description='CSS selector to limit search scope (e.g. "div#main")')
+ max_results: int = Field(default=25, description='Maximum matches to return')
+
+
+ class FindElementsAction(BaseModel):
+ selector: str = Field(description='CSS selector to query elements (e.g. "table tr", "a.link", "div.product")')
+ attributes: list[str] | None = Field(
+ default=None,
+ description='Specific attributes to extract (e.g. ["href", "src", "class"]). If not set, returns tag and text only.',
+ )
+ max_results: int = Field(default=50, description='Maximum elements to return')
+ include_text: bool = Field(default=True, description='Include text content of each element')
+
+
class SearchAction(BaseModel):
query: str
engine: str = Field(