IP Library Granted Patent US 12,578,839
Granted Patent B1
US 12,578,839 · App. 19/196,586 · Granted Mar 17, 2026

Graphical user interface for generating large language model function buttons

Inventors: Hiten Shah (Redwood City, CA); Richard Chan (Flushing, NY); Theo Richardson (Toronto, CA)
Assignee: Dropbox, Inc.
G06F3/04817G06F3/04842G06F9/451G06F40/186G06F40/221G06F40/30
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,578,839
App. No.
19/196,586
Granted
Mar 17, 2026
Kind
B1
Abstract

This disclosure describes systems that generate a selection of large language model (LLM) function buttons in a floating widget within a web browser of a client device. The disclosed systems can generate or otherwise select the LLM function buttons to include based on context of a webpage within the web browser. Responsive to detecting an indication of an interaction with an LLM function button, the disclosed systems can generate a side panel within the web browser according to the LLM function button. In some embodiments, the disclosed systems can display and utilize a customized LLM function button responsive to detecting a certain webpage or content within the webpage. Further, in some embodiments, the disclosed systems can generate an LLM function button to perform a customized workflow responsive to detecting a certain webpage or content within the webpage.

Claims (71)

1 . A computer-implemented method comprising:

receiving, from a client device displaying a webpage within a web browser, an indication of selecting a large language model (LLM) button generation option that is selectable to generate an LLM function button within the webpage that is selectable to autonomously generate a prompt to cause an LLM to autonomously perform a customized function based at least in part on content of the webpage;

in response to the indication of selecting the LLM button generation option, providing, for display within the web browser displayed on the client device, a button generation panel comprising fields to define parameters corresponding to the customized function of the LLM function button, the fields comprising:

an address option for defining one or more web addresses where the LLM function button appears; and

a function option for defining the customized function executed by the LLM upon selection of the LLM function button; and

generating, based on interactions with the fields of the button generation panel, the LLM function button within the webpage to be selectable to autonomously generate the prompt to cause the LLM to autonomously perform the customized function according to the parameters of the button generation panel.

2 . The computer-implemented method of claim 1 , further comprising defining the customized function executed upon selection of the LLM function button by:

receiving, from the client device displaying the webpage, an indication of selecting an action record option to record user account interaction with the webpage displayed within the web browser; and

in response to the indication of selecting the action record option, recording a set of user account interactions.

3 . The computer-implemented method of claim 2 , further comprising modifying the set of user account interactions to generate an adaptive set of user account interactions adaptable to differences in content within one or more webpages related to the webpage displayed within the web browser.

4 . The computer-implemented method of claim 3 , further comprising:

providing a description of the adaptive set of user account interactions for review in the button generation panel of the web browser;

generating a modified adaptive set of user account interactions by modifying the adaptive set of user account interactions responsive to feedback from the client device regarding the adaptive set of user account interactions; and

generating a prompt to provide to an LLM based on the modified adaptive set of user account interactions.

5 . The computer-implemented method of claim 1 , further comprising:

providing, within the button generation panel, a function adaptation option to adapt the customized function of the LLM function button to one or more webpages related to the webpage; and

adapting the customized function to the one or more webpages related to the webpage responsive to receiving an indication of an interaction with the function adaptation option.

6 . The computer-implemented method of claim 1 , further comprising displaying the LLM function button within a floating widget that overlays content of the webpage based on determining that a web address of the webpage matches the one or more web addresses where the LLM function button appears.

7 . The computer-implemented method of claim 1 , further comprising:

generating, based on detecting an indication of an interaction with the LLM function button, a side panel within the web browser; and

providing, within the side panel, a description of the customized function executed upon selection of the LLM function button.

8 . The computer-implemented method of claim 7 , further comprising:

executing the customized function responsive to receiving a confirmation from the client device to execute the customized function of the LLM function button.

9 . A system comprising:

at least one processor; and

at least one non-transitory computer readable storage medium storing instructions that, when executed by the at least one processor, cause the system to:

receive from a client device displaying a webpage within a web browser, an indication of selecting a large language model (LLM) button generation option that is selectable to generate an LLM function button within the webpage that is selectable to autonomously generate a prompt to cause an LLM to autonomously perform a customized function based at least in part on content of the webpage;

in response to the indication of selecting the LLM button generation option, provide, for display within the web browser displayed on the client device, a button generation panel comprising fields to define parameters corresponding to the customized function of the LLM function button, the fields comprising:

an address option for defining one or more web addresses where the LLM function button appears; and

a function option for defining the customized function executed by the LLM upon selection of the LLM function button; and

a function adaptation option to adapt the customized function of the LLM function button to one or more webpages related to the webpage; and

generate, based on interactions with the fields of the button generation panel, the LLM function button within the webpage to be selectable to autonomously generate the prompt to cause the LLM to autonomously perform the customized function according to the parameters of the button generation panel.

10 . The system of claim 9 , further comprising instructions that, when executed by the at least one processor, cause the system to:

receive via the function option, a natural language instruction defining the customized function executed upon selection of the LLM function button;

extract a workflow from the natural language instruction using a large language model, the workflow comprising a series of executable processes that, upon execution, accomplish the customized function defined via the function option; and

define the customized function according to the workflow extracted from the natural language instruction.

11 . The system of claim 9 , further comprising instructions that, when executed by the at least one processor, cause the system to:

provide, within the button generation panel, an interface element selectable to enable an LLM to access one or more content items stored within a content management system during execution of the customized function executed upon selection of the LLM function button.

12 . The system of claim 9 , further comprising instructions that, when executed by the at least one processor, cause the system to:

receive, from the client device displaying the webpage, an indication of selecting an action record option to record user account interaction with the webpage displayed within the web browser; and

in response to the indication of selecting the action record option, record a set of user account interactions.

13 . The system of claim 12 , further comprising instructions that, when executed by the at least one processor, cause the system to:

modify the set of user account interactions to generate an adaptive set of user account interactions adaptable to differences in content within one or more webpages related to the webpage displayed within the web browser.

14 . The system of claim 13 , further comprising instructions that, when executed by the at least one processor, cause the system to:

provide a description of the adaptive set of user account interactions for review in the button generation panel of the web browser;

generate a modified adaptive set of user account interactions by modifying the adaptive set of user account interactions responsive to feedback from the client device regarding the adaptive set of user account interactions; and

generate a prompt to provide to an LLM based on the modified adaptive set of user account interactions.

15 . A non-transitory computer-readable medium storing instructions thereon that, when executed by at least one processor, cause a computing device to:

receive, from a client device displaying a webpage within a web browser, an indication of selecting a large language model (LLM) button generation option that is selectable to generate an LLM function button within the webpage that is selectable to autonomously generate a prompt to cause an LLM to autonomously perform a customized function based at least in part on content of the webpage;

in response to the indication of selecting the LLM button generation option, provide, for display within the web browser displayed on the client device, a button generation panel comprising fields to define parameters corresponding to the customized function of the LLM function button, the fields comprising:

an address option for defining one or more web addresses where the LLM function button appears; and

a function option for defining the customized function executed by the LLM upon selection of the LLM function button; and

generate, based on interactions with the fields of the button generation panel, the LLM function button within the webpage to be selectable to autonomously generate the prompt to cause the LLM to autonomously perform the customized function according to the parameters of the button generation panel; and

display the LLM function button within a floating widget that overlays content of the webpage based on determining that a web address of the webpage matches the one or more web addresses where the LLM function button appears.

16 . The non-transitory computer-readable medium of claim 15 , further comprising instructions that, when executed by the at least one processor, cause the computing device to:

provide, within the button generation panel, a user interface element that is selectable to allow access to one or more content items within a content management system;

receive, via the function option, an instruction to extract data from the one or more content items; and

responsive to determining a selection of the LLM function button, extract the data according to the instruction.

17 . The non-transitory computer-readable medium of claim 15 , further comprising instructions that, when executed by the at least one processor, cause the computing device to:

generate, based on detecting an indication of an interaction with the LLM function button, a side panel within the web browser; and

provide, within the side panel, a description of the customized function executed upon selection of the LLM function button.

18 . The non-transitory computer-readable medium of claim 17 , further comprising instructions that, when executed by the at least one processor, cause the computing device to:

execute the customized function responsive to receiving a confirmation from the client device to execute the customized function of the LLM function button.

19 . The non-transitory computer-readable medium of claim 15 , further comprising instructions that, when executed by the at least one processor, cause the computing device to:

receive, from the client device displaying the webpage, an indication of selecting an action record option to record user account interaction with the webpage displayed within the web browser; and

in response to the indication of selecting the action record option, record a set of user account interactions.

20 . The non-transitory computer-readable medium of claim 19 , further comprising instructions that, when executed by the at least one processor, cause the computing device to:

modify the set of user account interactions to generate an adaptive set of user account interactions adaptable to differences in content within one or more webpages related to the webpage displayed within the web browser;

provide a description of the adaptive set of user account interactions for review in the button generation panel of the web browser;

generate a modified adaptive set of user account interactions by modifying the adaptive set of user account interactions responsive to feedback from the client device regarding the adaptive set of user account interactions; and

generate a prompt to provide to an LLM based on the modified adaptive set of user account interactions.

Assignments (2)
SECURITY INTEREST Recorded Sep 30, 2025
From: DROPBOX, INC.
To: WILMINGTON TRUST, NATIONAL ASSOCIATION, AS COLLATERAL AGENT
Reel/Frame 072425/0401 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded May 1, 2025
From: SHAH, HITEN; CHAN, RICHARD; RICHARDSON, THEO
To: DROPBOX, INC.
Reel/Frame 071002/0832 →
Continuity (1)
Provisional Application 63793486 · Apr 23, 2025
References Cited (40)
US 7953756B1 · Qin · 2011 [cited by applicant]
US 8463608B2 · Dow et al. · 2013 [cited by applicant]
US 8589950B2 · Treat et al. · 2013 [cited by applicant]
US 9916328B1 · Sharifi et al. · 2018 [cited by applicant]
US 11256735B2 · Narth et al. · 2022 [cited by applicant]
US 12321725B1 · Wilson et al. · 2025 [cited by applicant]
US 12380340B1 · Mamut et al. · 2025 [cited by applicant]
US 20020042688A1 · Banfer · 2002 [cited by applicant]
US 20060026532A1 · Pagan · 2006 [cited by examiner]
US 20120260203A1 · Commarford et al. · 2012 [cited by applicant]
US 20120268361A1 · Shotel et al. · 2012 [cited by applicant]
US 20170118258A1 · Lieb · 2017 [cited by applicant]
US 20170255865A1 · Riscutia et al. · 2017 [cited by applicant]
US 20180300028A1 · Vora et al. · 2018 [cited by applicant]
US 20190179875A1 · Lowery · 2019 [cited by examiner]
US 20210034693A1 · Semerad · 2021 [cited by examiner]
US 20210224818A1 · Choudhary et al. · 2021 [cited by applicant]
US 20210256076A1 · McMurray et al. · 2021 [cited by applicant]
US 20220012299A1 · Wittke et al. · 2022 [cited by applicant]
US 20240256615A1 · Liu et al. · 2024 [cited by applicant]
US 20240256618A1 · Adada · 2024 [cited by examiner]
US 20240281481A1 · Yushkina et al. · 2024 [cited by applicant]
US 20240320918A1 · Amador et al. · 2024 [cited by applicant]
US 20240362036A1 · Jacob et al. · 2024 [cited by applicant]
US 20240403328A1 · Hecht et al. · 2024 [cited by applicant]
US 20250005655A1 · Bennett et al. · 2025 [cited by applicant]
US 20250053430A1 · Jones et al. · 2025 [cited by applicant]
US 20250095808A1 · Davelaar et al. · 2025 [cited by applicant]
US 20250117573A1 · Azose et al. · 2025 [cited by applicant]
US 20250123863A1 · LaForge et al. · 2025 [cited by applicant]
US 20250173043A1 · Ma et al. · 2025 [cited by applicant]
US 20250199891A1 · Brazier et al. · 2025 [cited by applicant]
EP 2515513A1 · 2012 [cited by applicant]
EP 2853314A1 · 2015 [cited by applicant]
Non-Final Office Action from U.S. Appl. No. 19/196,588, mailed on Jul. 1, 2025, 22 pages. [cited by applicant]
Notice of Allowance for U.S. Appl. No. 19/196,582 mailed on Jul. 22, 2025, 5 pages. [cited by applicant]
Non-Final Office Action from U.S. Appl. No. 19/196,582, mailed on Jun. 5, 2025, 8 pages. [cited by applicant]
Final Office Action from U.S. Appl. No. 19/196,588 mailed on Sep. 10, 2025, 26 pages. [cited by applicant]
Notice of Allowance for U.S. Appl. No. 19/196,582 mailed on Oct. 21, 2025, 5 pages. [cited by applicant]
Notice of Allowance for U.S. Appl. No. 19/196,588 mailed on Nov. 18, 2025, 10 pages. [cited by applicant]
Cited By (1)
US 12,705,238