Skip to content

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

Fix param

Magnus Müller18b7dd95a4c2
diff --git a/browser_use/agent/system_prompt.md b/browser_use/agent/system_prompt.md
index b48a68cb6..fb58b6044 100644
--- a/browser_use/agent/system_prompt.md
+++ b/browser_use/agent/system_prompt.md
@@ -118,9 +118,9 @@ You must call the `done` action in one of two cases:
The `done` action is your opportunity to terminate and share your findings with the user.
- Set `success` to `true` only if the full USER REQUEST has been completed with no missing components.
- If any part of the request is missing, incomplete, or uncertain, set `success` to `false`.
-- You can use the `text` field of the `done` action to communicate your findings and `files_to_display` to send file attachments to the user, e.g. `["results.md"]`.
+- You can use the `text` field of the `done` action to communicate your findings and `files` to send file attachments to the user, e.g. `["results.md"]`.
- Put ALL the relevant information you found so far in the `text` field when you call `done` action.
-- Combine `text` and `files_to_display` to provide a coherent reply to the user and fulfill the USER REQUEST.
+- Combine `text` and `files` to provide a coherent reply to the user and fulfill the USER REQUEST.
- You are ONLY ALLOWED to call `done` as a single action. Don't call it together with other actions.
- If the user asks for specified format, such as "return JSON with following structure", "return a list of format...", MAKE sure to use the right format in your answer.
- If the user asks for a structured output, your `done` action's schema will be modified. Take this schema into account when solving the task!
diff --git a/browser_use/agent/system_prompt_no_thinking.md b/browser_use/agent/system_prompt_no_thinking.md
index c905e5bb6..6eddc625d 100644
--- a/browser_use/agent/system_prompt_no_thinking.md
+++ b/browser_use/agent/system_prompt_no_thinking.md
@@ -118,9 +118,9 @@ You must call the `done` action in one of two cases:
The `done` action is your opportunity to terminate and share your findings with the user.
- Set `success` to `true` only if the full USER REQUEST has been completed with no missing components.
- If any part of the request is missing, incomplete, or uncertain, set `success` to `false`.
-- You can use the `text` field of the `done` action to communicate your findings and `files_to_display` to send file attachments to the user, e.g. `["results.md"]`.
+- You can use the `text` field of the `done` action to communicate your findings and `files` to send file attachments to the user, e.g. `["results.md"]`.
- Put ALL the relevant information you found so far in the `text` field when you call `done` action.
-- Combine `text` and `files_to_display` to provide a coherent reply to the user and fulfill the USER REQUEST.
+- Combine `text` and `files` to provide a coherent reply to the user and fulfill the USER REQUEST.
- You are ONLY ALLOWED to call `done` as a single action. Don't call it together with other actions.
- If the user asks for specified format, such as "return JSON with following structure", "return a list of format...", MAKE sure to use the right format in your answer.
- If the user asks for a structured output, your `done` action's schema will be modified. Take this schema into account when solving the task!
diff --git a/browser_use/tools/service.py b/browser_use/tools/service.py
index 3e7b526a6..dbed5a1dc 100644
--- a/browser_use/tools/service.py
+++ b/browser_use/tools/service.py
@@ -683,19 +683,15 @@ You will be given a query and the markdown of a webpage that has been filtered t
# Look up the node from the selector map if index is provided
# Special case: index 0 means scroll the whole page (root/body element)
node = None
- if params.frame_idx is not None and params.frame_idx != 0:
- node = await browser_session.get_element_by_index(params.frame_idx)
+ if params.frame_index is not None and params.frame_index != 0:
+ node = await browser_session.get_element_by_index(params.frame_index)
if node is None:
# Element does not exist
- msg = f'Element index {params.frame_idx} not found in browser state'
+ msg = f'Element index {params.frame_index} not found in browser state'
return ActionResult(error=msg)
direction = 'down' if params.down else 'up'
- target = (
- 'the page'
- if params.frame_idx is None or params.frame_idx == 0
- else f'element {params.frame_idx}'
- )
+ target = 'the page' if params.frame_index is None or params.frame_index == 0 else f'element {params.frame_index}'
# Get actual viewport height for more accurate scrolling
try:
diff --git a/browser_use/tools/views.py b/browser_use/tools/views.py
index 692757c6f..c73752ba0 100644
--- a/browser_use/tools/views.py
+++ b/browser_use/tools/views.py
@@ -6,7 +6,7 @@ from pydantic import BaseModel, ConfigDict, Field
# Action Input Models
class SearchAction(BaseModel):
query: str
- search_engine: str = Field(
+ engine: str = Field(
default='duckduckgo', description='duckduckgo, google, bing (use duckduckgo by default because less captchas)'
)
@@ -39,7 +39,7 @@ class InputTextAction(BaseModel):
class DoneAction(BaseModel):
text: str = Field(description='summary for user')
success: bool = Field(description='True if user_request completed successfully')
- files_to_display: list[str] | None = Field(default=[], description='files to display')
+ files: list[str] | None = Field(default=[], description='files to display')
T = TypeVar('T', bound=BaseModel)