Granted Patent B2

US 12,591,607 · App. 18/216,922 · Granted Mar 31, 2026

Inventors: Myung Kyun Kim (Kirkland, WA); RadhaKrishna Hiremane (Seattle, WA); Thomas Andrew Albrecht (Minneapolis, MN); Joseph Andrew DeGregorio (Seattle, WA)

Assignees: ATLASSIAN PTY LTD.; ATLASSIAN US, INC.

G06F16/3334G06F3/04812G06F3/0486G06F9/451G06F9/541G06F9/547G06F16/243G06F16/2455G06F16/248G06F16/3329G06F16/345G06F16/90332G06F16/9038G06F16/93G06F16/954G06F21/31G06F40/117G06F40/134G06F40/166G06F40/174G06F40/186G06F40/197G06F40/20G06F40/205G06F40/30G06F40/40G06N3/0475G06Q10/063114G06Q10/06316G06Q10/101H04L51/02H04L51/21G06F3/0484G06Q10/103

Abstract

Claims (115)

1 . A computer-implemented method for providing technical assistance in respect of a collaboration platform using a messaging platform separate from the collaboration platform, the method comprising:

receiving a request message generated at a first client device by a first user, the request message including a natural language user input referencing a technical problem in respect of the collaboration platform and provided to a chat interface rendered by the messaging platform;

computing a completeness score using the natural language user input;

in response the completeness score failing to satisfy a criteria, automatically generating a first prompt comprising:

first predetermined query prompt text; and

at least a portion of the natural language user input;

communicating the first prompt to a generative output engine separate from the messaging platform;

receiving a first response from the generative output engine, the response comprising a suggested request;

causing the suggested request to be transmitted to and rendered by the first client device via the messaging platform; and

conducting a search of a knowledge base platform separate from the messaging platform using at least a portion of the natural language user input or a subsequent natural language user input provided subsequent to the suggested request being transmitted;

in response to receiving a set of search results in response to the search of the knowledge base platform, each result identifying a respective knowledge base content item, generating a second prompt comprising:

second predetermined query prompt text; and

content extracted from at least one knowledge base content item identified in the set of search results;

communicating the second prompt to the generative output engine;

receiving a second response from the generative output engine; and

generating a response message comprising:

a selectable link to the knowledge base platform for each of at least a subset of the set of search results;

a confidence score computed using the natural language user input and content extracted from the at least one knowledge base content item;

a summary of the content extracted form the at least one knowledge base content item extracted from the second response; and

at least a portion of the subset of the set of search results; and

providing the response message for transmission to and rendering on the first client device via the messaging platform.

2 . The computer-implemented method of claim 1 , wherein the method further comprises:

for each search result to be included in the response message, generating a respective prompt comprising:

the second predetermined query prompt text; and

respective content extracted from each respective knowledge base content item identified in the set of search results that corresponds to the result to be included in the response message; and

receiving a respective response from the generative output engine for each respective prompt.

3 . The computer-implemented method of claim 2 , wherein each respective response is positioned proximate to a corresponding selectable link in the response message.

4 . The computer-implemented method of claim 2 , wherein the method further comprises:

computing a confidence score for each result to be included in the response message; and

including each confidence score proximate to the corresponding selectable link in the response message.

5 . The computer-implemented method of claim 1 , wherein the method further comprises:

extracting content from one or more knowledge base content items identified in the set of search results; and

generate a third prompt comprising:

third predetermined query prompt text; and

the content extracted from the one or more knowledge base content items identified in the set of search results;

communicate the third prompt to the generative output engine;

receive a third response from the generative output engine; and

generate a second response message comprising the third response.

6 . The computer-implemented method of claim 5 , wherein:

the third predetermined query prompt text includes a request for a proposed list of suggested operations; and

the second response message includes at least a list of steps for the user provided by the generative output engine.

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

the first predetermined query prompt text includes a request for an expanded request; and

the suggested request comprises at least one additional inquiry as compared to the natural language user input.

8 . The computer-implemented method of claim 1 , wherein the subsequent natural language user input is the suggested request.

9 . The computer-implemented method of claim 1 , wherein:

the first prompt is communicated to the generative output engine using an application programming interface (API) call transmitted to the generative output engine;

the first prompt is a first serialized data object provided to the generative output engine using the API call;

the second prompt is communicated to the generative output engine using the API call; and

the second prompt is provided as a second serialized data object to the generative output engine using the API call.

10 . A system for providing content for a chat session with a user, the system comprising:

a backend of a collaboration platform configured to provide technical assistance, the backend of the collaboration platform executing on a server having a processing unit and computer readable memory operably coupled to the processing unit and storing instructions that, when accessed by the processing unit, configure the backend of the collaboration platform for:

receiving a request message generated at a first client device by the user, the request message including a natural language user input referencing a technical problem in respect of the collaboration platform and provided to a chat interface of a messaging platform separate from the collaboration platform;

computing a question completeness score using the natural language user input;

in response to the question completeness score failing to satisfy a criteria, generating a first prompt comprising:

first predetermined query prompt text; and

at least a portion of the natural language user input;

automatically communicating the first prompt to a generative output engine;

receiving a first response from the generative output engine, the response comprising a suggested question for eliciting further information from the user;

causing the suggested question to be transmitted to the first client device via the messaging platform; and

in response to a user input provided to the first client device after the suggested question is transmitted to the client device, conducting a search of a content platform separate from the messaging platform using at least a portion of the user input;

in response to receiving a set of search results from the search of the content platform, each result identifying a respective content item, generating a second prompt comprising:

second predetermined query prompt text; and

content extracted from at least one content item identified in the set of search results;

communicating the second prompt to the generative output engine;

receiving a second response from the generative output engine;

generating a response message comprising:

the second response from the generative output engine;

a selectable link to the content platform to one or more of the content items identified in the set of search results;

a confidence score computed using the natural language user input and content extracted from the respective content item

a summary of the content extracted form the at least one content item extracted; and

providing the response message for transmission to the first client device via the messaging platform.

11 . The system of claim 10 , wherein the instructions further comprise instructions for:

extracting content from one or more content items identified in the set of search results; and

generating a third prompt comprising:

third predetermined query prompt text; and

the content extracted from the one or more content items identified in the set of search results;

communicating the third prompt to the generative output engine;

receiving a third response from the generative output engine; and

generating a second response message comprising the third response.

12 . The system of claim 11 , wherein:

the third predetermined query prompt text includes a request for a proposed list of suggested operations; and

the second response message includes at least a list of steps for the user generated by the generative output engine.

13 . A computer-implemented method for providing technical assistance using a messaging platform, the method comprising:

receiving a request message generated at a first client device by a first user, the request message including a natural language user input referencing a technical problem in respect of a collaboration platform, the natural language user input provided to a chat interface of a messaging platform separate from the collaboration platform;

in response to receiving the request message, computing a question completeness score;

in response to the question completeness score failing to satisfy a criteria, generate a first prompt comprising:

first predetermined query prompt text; and

at least a portion of the natural language user input;

communicating the first prompt to a generative output engine;

receiving a first response from the generative output engine, the response comprising a list of suggested additional questions;

causing the list of suggested additional questions to be transmitted to the first client device via the messaging platform;

in response to a user input provided to the first client device after the list of suggested additional question, conducting a search of a content platform using the user input;

in response to receiving a set of search results from the search of the content platform, each result identifying a respective content item, generating a second prompt comprising:

second predetermined query prompt text; and

content extracted from at least one content item identified in the set of search results;

communicating the second prompt to the generative output engine;

receiving a second response from the generative output engine, the second response comprising a confidence score based at least in part on the natural language user input and the content extracted from the at least one content item, a link to the at least one content item and a summary generated in respect of the content extracted from the at least one content item;

generating a response message comprising the second response from the generative output engine; and

providing the response message for transmission to the first client device via the messaging platform.

14 . The computer-implemented method of claim 13 , wherein:

the response message further comprises:

a selectable link to each of a subset of the set of search results; and

a content summary positioned proximate to each selectable link; and

each respective content summary is generated using the generative output engine.

15 . The computer-implemented method of claim 14 , wherein:

the content summary is computed using a portion of text extracted from the respective content item; and

the portion of text is identified using a semantic analysis of the content item as compared to the natural language user input text.

16 . The computer-implemented method of claim 13 , wherein the method further comprises:

subsequent to transmitting the response message to the first client device, requesting feedback with respect to content items identified in the response message; and

in response to receiving a negative feedback with respect to a particular content item, causing the particular content item to be suppressed from subsequent search results.

17 . The computer-implemented method of claim 13 , wherein the response message is transmitted to an operator before being relayed to the first client device via the messaging platform.

18 . The computer-implemented method of claim 13 , wherein:

the first prompt is communicated to the generative output engine using an API call transmitted to the generative output engine; and

the second prompt is communicated to the generative output engine using the API call.

Assignments (1)

Continuity (2)

References Cited (67)