CTFR 18/670,398 CTFR 90841 DETAILED ACTION Remarks Claims 1-20 have been examined and rejected. This Office action is responsive to the amendment filed on 05/26/2026, which has been entered in the above identified application. Notice of Pre-AIA or AIA Status 07-03-aia AIA 15-10-aia The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA. Claim Objections 07-29-01 AIA Claim 13, 14, and 18 are objected to because of the following informalities: Claims 13 and 14 recite ‘the one or more execution examples from the set of execution examples’; however, it should recite - - the one or more execution examples selected from the set of execution examples - -. Claim 18 recites ‘from the estimated lookahead plan using and one or more large language models’; however, it should recite - - from the estimated lookahead plan and using one or more large language models - -. Appropriate correction is required. Claim Rejections - 35 USC § 112 07-30-02 AIA The following is a quotation of 35 U.S.C. 112(b): (b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention. The following is a quotation of 35 U.S.C. 112 (pre-AIA), second paragraph: The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention. 07-34-01 Claim 7 is rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention. Regarding claim 7 , claim 7 recites “the estimated lookahead plan describing the one or more actions to be executed after the set of previous actions in performing the task”. The claim lacks antecedent basis for “the one or more actions to be executed after the set of previous actions in performing the task”. It is unclear how this relates the previously recited “the estimated lookahead plan describing the one or more actions for performing the task”. For the purposes of examination, this limitation is interpreted as: a second estimated lookahead plan describing the one or more actions to be executed after the set of previous actions in performing the task”. Claim Rejections - 35 USC § 101 07-04-01 AIA 07-04 35 U.S.C. 101 reads as follows: Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title. Claims 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. Claims 1, 10, and 18 Step 1 : Claims 1, 10, and 18 recite a method, a system, and a medium; therefore, they are directed to the statutory categories of a method, a machine, and a manufacture. Step 2A Prong 1 : Claim 1 recites, inter alia: generating a lookahead prompt comprising at least one execution example corresponding to the task, the at least one execution example including an example task and an example action sequence for performing the example task; Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of creating a prompt including an example, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper. generating, from the lookahead prompt, an estimated lookahead plan describing one or more actions for performing the task; Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of creating a plan from the prompt, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper. generating, from the estimated lookahead plan, instructions to perform a next action in a sequence for performing the task, the instructions indicating an interactive element selected from the set of interactive elements; Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of creating instructions to perform an action, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper. Claim 10 recites, inter alia: determine, from the set of interactive elements of the software application, a set of candidate interactive elements that correspond to performance of the task; generate, from the set of candidate interactive elements, an estimated lookahead plan describing one or more actions for performing the task; Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of creating a plan for performing a task, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper. generate, from the set of candidate interactive elements, an operation for a next action in a sequence for performing the task; Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of creating an operation for an action of a task, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper. determine, from the set of candidate interactive elements and from the estimated lookahead plan, an interactive element from the set of candidate interactive elements to target via the operation of the next action; Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of determining an element to target based on a plan, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper. generate instructions that indicate the interactive element; Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of creating instructions to perform an action, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper. Claim 18 recites, inter alia: determining an environment representation; determining one or more previous actions that have been executed in performance of the task; Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of determining an environment, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper. generating an estimated lookahead plan describing one or more remaining actions for performing the task by generating the estimated lookahead plan based on the environment representation, the one or more previous actions, and an execution example that includes an example task and an example action sequence for performing the example task; Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of creating a plan describing actions for a task, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper. generating, from the estimated lookahead plan, instructions to perform a next action for performing the task via user interaction with an interactive element, the instructions indicating an interactive element selected from the set of interactive elements; Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of creating instructions to perform an action from the plan, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper. Step 2A Prong 2 : This judicial exception is not integrated into a practical application. The additional elements of “ A computer-implemented method comprising ”, “ A system comprising: one or more memory devices; and one or more processors configured to cause the system to ”, “ A non-transitory computer-readable medium storing executable instructions which, when executed by a processing device, cause the processing device to perform operations comprising ”, “ using a large language model ”, “ using one or more large language models ”, “ via user interaction with an interactive element of the user interface ”, “ a set of candidate interactive elements of the user interface of the software application ” “ using a first large language model ”, “ using a second large language model ”, “ of the user interface ”, “ using the first large language model ”, “ the one or more actions involving use of one or more interactive elements from the set of interactive elements ”, and “ wherein the one or more remaining actions involve use of one or more interactive elements from the set of interactive elements ” amount to no more than generally linking the use of a judicial exception to a particular technological environment or field of use (see MPEP § 2106.05(h). The claimed computer components are recited at a high level of generality and are merely invoked as tool to perform the abstract idea. The additional elements of “receiving, from a client device interacting with a software application, a query regarding a task to be performed via a user interface of the software application, the software application having a set of interactive elements that are accessible via the user interface “, “ the instructions indicating an interactive element selected from the set of interactive elements and prompting user interaction with the interactive element via the user interface in performance of the next action ”, and “ instructions that indicate the interactive element and prompt user interaction with the interactive element via the user interface in execution of the operation ” amount to insignificant extra-solution activity in the form of mere data gathering and output (see MPEP § 2106.05(g)). Even when viewed in combination, these additional element do not integrate the abstract idea into a practical application and the claims are thus directed to the abstract idea. Step 2B : The claims do not contain significantly more than the judicial exception. “ A computer-implemented method comprising ”, “ A system comprising: one or more memory devices; and one or more processors configured to cause the system to ”, “ A non-transitory computer-readable medium storing executable instructions which, when executed by a processing device, cause the processing device to perform operations comprising ”, “ using a large language model ”, “ using one or more large language models ”, “ via user interaction with an interactive element of the user interface ”, “ a set of candidate interactive elements of the user interface of the software application ” “ using a first large language model ”, “ using a second large language model ”, “ of the user interface ”, “ using the first large language model ”, “ the one or more actions involving use of one or more interactive elements from the set of interactive elements ”, and “ wherein the one or more remaining actions involve use of one or more interactive elements from the set of interactive elements ” amount to no more than generally linking the use of a judicial exception to a particular technological environment or field of use (see MPEP § 2106.05(h)). Mere instructions to apply an exception using a generic computer component cannot provide an inventive concept. The additional elements of “receiving, from a client device interacting with a software application, a query regarding a task to be performed via a user interface of the software application, the software application having a set of interactive elements that are accessible via the user interface“, “the instructions indicating an interactive element selected from the set of interactive elements and prompting user interaction with the interactive element via the user interface in performance of the next action”, and “instructions that indicate the interactive element and prompt user interaction with the interactive element via the user interface in execution of the operation” amounts to insignificant extra-solution activity in the form of mere data gathering and output (see MPEP § 2106.05(g)), and is a well-understood, routine, conventional activity (see MPEP § 2106.05(d); “Receiving or transmitting data over a network”). Nothing in the claims provides significantly more than that abstract idea. As such, the claims are ineligible. Claims 2-9, 11-17, 19, and 20 Step 1 : Claims 2-9, 11-17, 19, and 20 recite a method, a system, and a medium; therefore, they are directed to the statutory categories of a method, a machine, and a manufacture. Step 2 : claims 2-9, 11-17, 19, and 20 merely narrow the previously recited abstract idea limitations. For the reasons described above with respect to claims 1, 10, and 18, this judicial exception is not meaningfully integrated into a practical application, or significantly more than the abstract idea. The claims disclose similar limitations described for the independent claims above and do not provide anything more than the mental processes that are practically capable of being performed in the human mind with the assistance of pen and paper and mathematical concepts that are achievable through mathematical computation. Claim 2 further recites the additional element of “ wherein generating the instructions to perform the next action in the sequence for performing the task comprises generating an operation to execute in performing the next action ”. Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of creating an operation for an action, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper. The additional elements of “ using the one or more large language models ” and “ using an additional large language model ” amount to no more than generally linking the use of a judicial exception to a particular technological environment or field of use (see MPEP § 2106.05(h)). Claim 3 further recites the additional element of “ wherein generating the instructions to perform the next action in the sequence for performing the task comprises determining to target the interactive element selected from the set of interactive elements via the operation to execute in performing the next action” Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of determining to target an element, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper. The additional elements of “ using the one or more large language models ”, “ using the large language model ”, and amount to no more than generally linking the use of a judicial exception to a particular technological environment or field of use (see MPEP § 2106.05(h)). Claim 4 further recites the additional element of “ further comprising generating a target element prompt that includes chain-of-thought reasoning, the chain-of-thought reasoning related to selection of a target interactive element based on a determined operation; wherein determining to target the interactive element via the operation comprises determining to target the interactive element by incorporating the chain-of-thought reasoning of the target element prompt ”. Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of generating a prompt including reasoning and determining to target an element based on reasoning, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper. The additional elements of “ via the large language model ” amount to no more than generally linking the use of a judicial exception to a particular technological environment or field of use (see MPEP § 2106.05(h)). Claim 5 further recites the additional element of “ generating an encoding of the query; generating a plurality of additional encodings for a plurality of execution examples that are from a dataset of execution examples; determining pairwise cosine similarities between the encoding of the query and the plurality of additional encodings; and selecting the at least one execution example from the plurality of execution examples to include in the lookahead prompt based on the pairwise cosine similarities ”. Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of encoding examples and selecting an example based on cosine similarities, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper, or is a mathematical concept that is achievable through mathematical computation. The additional elements of “ using an encoding model ” and “ using the encoding model ” amount to no more than generally linking the use of a judicial exception to a particular technological environment or field of use (see MPEP § 2106.05(h)). Claim 6 further recites the additional element of “ generating the plurality of additional encodings for the plurality of execution examples comprises generating the plurality of additional encodings for a set of execution examples; and selecting the at least one execution example to include in the lookahead prompt based on the pairwise cosine similarities comprises selecting the at least one execution example based on the pairwise cosine similarities and based on determining that the at least one execution example ”. Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of encoding examples and selecting an example based on cosine similarities, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper, or is a mathematical concept that is achievable through mathematical computation. The additional elements of “ associated with a plurality of software applications ” and “ is associated with the software application ” amount to no more than generally linking the use of a judicial exception to a particular technological environment or field of use (see MPEP § 2106.05(h)). Claim 7 further recites the additional element of “ further comprising determining a set of previous actions that have been executed in performance of the task, wherein generating the estimated lookahead plan describing the one or more actions for performing the task comprises generating the estimated lookahead plan describing the one or more actions to be executed after the set of previous actions in performing the task ”. Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of determining previous actions and generating a plan describing actions, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper. Claim 8 further recites the additional element of “ further comprising determining an environment representation, wherein generating the lookahead prompt comprising the at least one execution example corresponding to the task comprises generating the lookahead prompt comprising the at least one execution example and the environment representation ”. Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of determining an environment representation and determining a prompt based on an example the environment representation, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper. The additional elements of “ of the user interface ” amount to no more than generally linking the use of a judicial exception to a particular technological environment or field of use (see MPEP § 2106.05(h)). Claim 9 further recites the additional element of “ wherein determining the environment representation of the user interface comprises determining Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of determining an environment representation, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper. The additional elements of “ a hypertext markup language representation of the user interface”. ” amount to no more than generally linking the use of a judicial exception to a particular technological environment or field of use (see MPEP § 2106.05(h)). Claim 11 further recites the additional element of “ determine the set of candidate interactive elements: generating from the query and from one or more previous actions that have been executed in performance of the task, a ranking of the plurality of interactive elements; and determining the set of candidate interactive elements from the plurality of interactive elements based on the ranking. ”. Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of generating a ranking and determining elements from the ranking, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper, or is a mathematical concept that is achievable through mathematical computation. The additional elements of “ wherein the one or more processors are configured to cause the system to ”, “ of the user interface ”, and “ using a ranking model ” amount to no more than generally linking the use of a judicial exception to a particular technological environment or field of use (see MPEP § 2106.05(h)). The additional elements of “ extracting, from an environment representation, a plurality of interactive elements “ amount to insignificant extra-solution activity in the form of mere data gathering and output (see MPEP § 2106.05(g)). Claim 12 further recites the additional element of “ generate the estimated lookahead plan from the set of candidate interactive elements using the first large language model by: determining a set of execution examples corresponding to the task, each execution example from the set of execution examples including an example task and an example action sequence for performing the example task; generating pairwise cosine similarities between an encoding of the query and a plurality of additional encodings of the set of execution examples; and generating the estimated lookahead plan from the set of candidate interactive elements and one or more execution examples selected from the set of execution examples, wherein the one or more execution examples are selected based on the pairwise cosine similarities ”. Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of determining examples, determining similarities, and creating a plan based on the similarities, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper, or is a mathematical concept that is achievable through mathematical computation. The additional elements of “ wherein the one or more processors are configured to cause the system to ” and “ using the first large language model ” amount to no more than generally linking the use of a judicial exception to a particular technological environment or field of use (see MPEP § 2106.05(h)). Claim 13 further recites the additional element of “ select the one or more execution examples from the set of execution examples based on the pairwise cosine similarities by: determining that one or more pairwise cosine similarities determined for the one or more execution examples satisfy a pairwise cosine similarity threshold; or determining that the one or more pairwise cosine similarities determined for the one or more execution examples indicate a higher similarity to the query than remaining execution examples from the set of execution examples ”. Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of comparing similarities, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper, or is a mathematical concept that is achievable through mathematical computation. The additional elements of “ wherein the one or more processors are configured to cause the system to ” amount to no more than generally linking the use of a judicial exception to a particular technological environment or field of use (see MPEP § 2106.05(h)). Claim 14 further recites the additional element of “ cause the system to select the one or more execution examples from the set of execution examples based on the pairwise cosine similarities by determining that the one or more execution examples are ”. Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of encoding examples and selecting examples, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper. The additional elements of “ wherein the one or more processors are configured to cause the system to ” and “ associated with the software application ” amount to no more than generally linking the use of a judicial exception to a particular technological environment or field of use (see MPEP § 2106.05(h)). Claim 15 further recites the additional element of “ determine the interactive element from the set of candidate interactive elements and the estimated lookahead plan by: determining a set of target element examples, each target element example including an example task and an example interactive element, the example interactive element to be targeted by an example operation in performance of an example next action for the example task; and determining, from the set of candidate interactive elements, the interactive element based on the estimated lookahead plan and the set of target element examples ”. Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of determining examples and determining elements, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper. The additional elements of “ wherein the one or more processors are configured to cause the system to ” and “ using the first large language model ” amount to no more than generally linking the use of a judicial exception to a particular technological environment or field of use (see MPEP § 2106.05(h)). Claim 16 further recites the additional element of “ wherein determining, from the set of candidate interactive elements, the interactive element based the estimated lookahead plan and the set of target element examples comprises: generating a target element prompt that includes chain-of-thought reasoning, the chain-of-thought reasoning related to selection of a target interactive element based on a determined operation, the chain-of-thought reasoning incorporating the set of candidate interactive elements, the estimated lookahead plan, and the set of target element examples; and determining, based on the target element prompt, the interactive element from the target element prompt ”. Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of creating a prompt including reasoning and determining an element, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper. The additional elements of “ using the first large language model ” amount to no more than generally linking the use of a judicial exception to a particular technological environment or field of use (see MPEP § 2106.05(h)). Claim 17 further recites the additional element of “ generate the operation for the next action in the sequence for performing the task by determining that the operation includes a type operation or a select operation; and determine a value as an argument for the type operation or the select operation ”. Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of determining an operation type and determining a value, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper. The additional elements of “ wherein the one or more processors are configured to cause the system to ” amount to no more than generally linking the use of a judicial exception to a particular technological environment or field of use (see MPEP § 2106.05(h)). Claim 19 further recites the additional element of “ wherein generating, from the estimated lookahead plan, the instructions to perform the next action comprises generating, from a next operation indicated by the estimated lookahead plan, the instructions to perform the next action ”. Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of creating instructions for an action, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper. The additional elements of “ using the one or more large language models ” amount to no more than generally linking the use of a judicial exception to a particular technological environment or field of use (see MPEP § 2106.05(h)). Claim 20 further recites the additional element of “ wherein generating, from the next operation indicated by the estimated lookahead plan, the instructions to perform the next action comprises: generating an operation for the next action; and generating the instructions to include the operation as part of the next action based on comparing the operation generated to the next operation indicated by the estimated lookahead plan ”. Under its broadest reasonable interpretation in light of the specification, this limitation encompasses the mental process of generating an operation and creating instructions based on comparing operations, which is an evaluation or observation that is practically capable of being performed in the human mind with the assistance of pen and paper. The additional elements of “ using the one or more large language models ”, “ using an additional large language model ”, and “ via the additional large language model ” amount to no more than generally linking the use of a judicial exception to a particular technological environment or field of use (see MPEP § 2106.05(h)). Claim Rejections - 35 USC § 103 07-06 AIA 15-10-15 In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status. 07-20-aia AIA The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action: A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made. 07-21-aia AIA Claim s 1-4, 7, 8, 10, 11, and 15-20 are rejected under 35 U.S.C. 103 as being unpatentable over Fabian et al. (US 20240303422 A1, published 09/12/2024), hereinafter Fabian, in further view of Mathur et al. (US 20240412226 A1, published 12/12/2024), hereinafter Mathur . Regarding claim 1, Fabian in view of Mathur teaches the claim comprising: A computer-implemented method comprising: receiving, from a client device interacting with a software application, a query regarding a task to be performed via a user interface of the software application, the software application having a set of interactive elements that are accessible via the user interface (Fabian Figs 1-13; [0082], FIGS. 5B-5F continue operational scenario 500, illustrating a turn-based chat including inputs received from a user, prompts generated based on the inputs, and responses based on replies to the inputs from LLM 330. In FIG. 5B, the user enters the natural language inquiry about data table 502 into task pane 503. User interface 307 transmits the user input to prompt engine 305 which generates a prompt based on the input; [0099-0101], FIGS. 8A, 8B, and 8C illustrate operational scenario 800 which presents an alternative presentation of the reply to the second prompt from LLM 330 beginning with, in FIG. 8A, receiving the user input requesting a way to calculate the surface area of a cylinder; [0108], FIGS. 10A, 10B, and 10C illustrate operational scenario 1000 of an LLM integration with a spreadsheet environment) ; generating a lookahead prompt comprising at least one execution example corresponding to the task, the at least one execution example including an example task and an example action sequence for performing the example task (Fabian Figs 1-13; [0032], To generate a prompt based on a user's natural language input, the application will configure the prompt according to one or more of: the scope of the problem, the tasks to be completed, illustrative examples, sample data or spreadsheet contextual information, rules, the output format, and a cue to get the LLM to complete the task and not ramble. The scope of the problem refers to the domain in which the problem is to be addressed. For example, the prompt may instruct the LLM to limit its reply to Microsoft Excel® formulas and not Google Sheets formulas. The tasks to be completed include instructions, such as, “generate a formula to do X and self-evaluate the formula.” Illustrative examples can include an example of how to self-evaluate a formula when the formula is relatively complex. An illustrative example provides the LLM with guidance on how to complete a task of the prompt; [0060], Application service 110 receives the input and generates a prompt to be submitted to LLM service 120 based on the input. The prompt includes contextual information, such as a chat history and spreadsheet data or metadata, such as a table name, worksheet name, or spreadsheet name, row and column headers, and at least a subset of the data, e.g., the first five or ten rows of data. Contextual information can also include the recent or latest actions performed by the user on the spreadsheet, such as the user creating a new column or entering a formula. Contextual information can also include errors detected in the spreadsheet; [0066], To configure the prompt, prompt engine 305 identifies a prompt template according to the type of request in the input in an implementation. Prompt templates can include prompt configurations for suggesting a calculated column to be added to workbook data 320, for a general inquiry about workbook data 320, for analyzing data in workbook data 320 to project a result, such as for a hypothetical scenario, and so on; [0067], Prompt engine 305 configures the prompt including parameters to direct LLM 330 to provide a focused response to the input. Prompt parameters include the scope of the prompt, the output format of the reply to produce a reply in a parse-able format, instructions or tasks, examples including sample data or sample data formatting, special tokens which influence the behavior of LLM 330, and so on; [0082], User interface 307 transmits the user input to prompt engine 305 which generates a prompt based on the input. Prompt engine 305 selects a prompt template based on a type or classification of the inquiry and configures a prompt according to the template. In the prompt, prompt engine 305 includes tasks, instructions, or rules applicable to generating a reply to the input, such as tasking LLM 330 to perform a self-evaluation of its reply and a rule to return the reply in a particular output format which is suitable for parsing; [0083], Prompt engine 305 also includes contextual information in the prompt. Prompt engine 305 retrieves spreadsheet context data relating to data table 502 from various ones of application components 303. For example, for a general inquiry about the contents of data table 502, prompt engine 305 configures the prompt to include all the column headers in data table 502 along with the table name and the first five rows of data) ; generating, from the lookahead prompt and using a large language model, an estimated lookahead plan describing one or more actions for performing the task, the one or more actions involving use of one or more interactive elements from the set of interactive elements; and generating, from the estimated lookahead plan and using one or more large language models, instructions to perform a next action in a sequence for performing the task, the instructions indicating an interactive element selected from the set of interactive elements (Fabian Figs 1-13; [0028], the reply from the LLM includes a specific suggestion for modifying the spreadsheet in the specific way; The LLM, upon receiving the configured prompt, generates a reply which suggests adding a column which computes a total sales quantity based on year-to-date (YTD) sales data along with a spreadsheet formula to calculate the quantity. Based on the reply, the application displays the suggestion along with the formula in the user interface of the spreadsheet. In an implementation, the application generates a preview of the column to be added to the data table based on the suggestion, i.e., a column calculating YTD sales data based on the table data and using the formula provided in the reply of the LLM, which the user may then direct the application to implement; [0048], Application service 110 may display the suggestions according to how LLM service 120 self-evaluated the suggestions, e.g., according to relevance to the input or correctness; [0049], Upon receiving the user's selection of the first suggestion in task pane 144, application service 110 implements the suggestion by adding a column (not shown) to the spreadsheet data; [0084], Continuing with FIG. 5B, with the prompt configured, prompt engine 305 sends the prompt to LLM 330. LLM 330 generates a reply to the prompt and transmits the reply to prompt engine 305; [0087], Upon receiving LLM 330's reply, prompt engine 305 post-processes the reply to generate a response for display in user interface 307. Post-processing the reply includes extracting information from the reply according to the output formatting rules, such as a description of the suggestion and instructions by which the suggestion can be implemented either by application components 303 or by the user. Continuing to FIG. 5D, the response includes a link by which the user can view more detailed information about the suggestion and a graphical button by which to add the column to data table 502. The user clicks the “Add column” button, causing one or more of application components 303 to modify data table 502 to include the new column and to fill the column according to instructions provided in the reply; [0088-0089], LLM 330 generates two formulas along with the descriptions and explanations and determines, in FIG. 5F, that the formulas are of moderate and high relevance to the input. LLM 330 transmits its reply to prompt engine 305 which, in turn, generates a response and sends the response to user interface 307 for display; [0094], the application service inputs a second prompt to the LLM service which instructs the LLM service to provide a chain-of-thought breakdown or decomposition of the formula generated for the first prompt; [0095], the second prompt generated by the application service directs the LLM service to return the formula generated in response to the first prompt in terms of the Microsoft Excel LET function; [0097], FIGS. 7A and 7B illustrate operational scenario 700 of prompting an LLM service by an application service involving the use of chain-of-thought decomposition; [0098], In FIG. 7B, upon receiving the formula in response to the first prompt, prompt engine 305 generates a second prompt which tasks LLM 330 with using Excel's LET function to break the formula down into steps; [0099-0101], FIGS. 8A, 8B, and 8C illustrate operational scenario 800 which presents an alternative presentation of the reply to the second prompt from LLM 330 beginning with, in FIG. 8A, receiving the user input requesting a way to calculate the surface area of a cylinder; Based on the reply from LLM 330 to the second prompt, prompt engine 305 configures and progressively displays the parts or steps of the output in user interface 307. As each part is displayed, the user can manually advance to the disclosure of the next part, for example, by clicking an “Accept.”; [0102], FIGS. 9A-9E illustrate operational scenario 900; [0103], In FIG. 9A, the user selects a cell next to the last column of the data table. The application floats a suggestion to allow the application to create a column for the user; [0107], FIG. 9D illustrates task pane 910 subsequent to the user selecting the Years of Service column for inclusion in the data table. In textbox 911, the user enters a natural language input to “Add a column with the employee's last name.” The application configures a prompt based on the input which tasks the LLM to produce suggestions in response to the input; [0108], FIGS. 10A, 10B, and 10C illustrate operational scenario 1000 of an LLM integration with a spreadsheet environment; [0109], The user is also presented with textbox 1013 by which the user can submit natural language inputs, such as requests or queries; [0110], The LLM generates and sends a reply to the prompt to the application which the application processes to produce a response to the input. The response to the input from the application based on the reply from the LLM is shown in output 1015. Output 1015 includes an explanation of the reply and a table demonstrating a spreadsheet calculation, including spreadsheet formulas, demonstrating a calculation responsive to the input; [0111], In FIG. 10C, task pane 1010 is updated by the application to show input 1016 and output 1017 based on the LLM's reply to the prompt based on input 1016. Output 1017 includes three suggestions received from the LLM in response to input 1016 according to parameters provided by the application in the prompt) However, Fabian in view of Mathur fails to expressly disclose the instructions indicating an interactive element selected from the set of interactive elements and prompting user interaction with the interactive element via the user interface in performance of the next action. In the same field of endeavor, Mathur teaches: the instructions indicating an interactive element selected from the set of interactive elements and prompting user interaction with the interactive element via the user interface in performance of the next action (Mathur Figs. 1-7; [0015], the response may miss certain steps that are crucial information for the user in performing an action; the response may need to include a sequential number of steps for performing an action in the application; [0021], The application 120 and local applications 164A-164N (collectively referred to as local application 164) may be any application that enables a user such as users 162A-162N (collectively referred to as user 162) to interact with the application to perform an action or achieve a purpose; [0024], the application 120 and/or application 164 utilize the LLM 130 to provide an application copilot that assists users in navigating application features and/or performing actions within the application; [0036], The path extraction engine 240 examines the response to extract a relevant path of actions included in the response. The terms “action path”, “path of actions” or path as used herein refer to a sequence of terms included in a response from the language model or in a help documentation, where each term refers to an action a user takes or refers to a user interface elements a user selects to take an action to perform a task associated with the product help inquiry., the extracted action path is a sequence of terms used in the response that correspond with specific actions or user interface elements in the product. For example, the action path for the example response depicted in FIG. 4B which explains how to create a new calendar meeting, is “New Meeting”, “Invite People”, and “Send.” The actions correspond with specific user interface elements of the meeting application. The path extraction engine 240 may utilize information about the product/application to retrieve key terms in the response that are associated with product actions/UI elements; [0044], FIGS. 4A-4C depict example user interface (UI) screens of an application that utilizes a language model to provide responses to product help inquiries. FIG. 4A depicts an example graphical user interface (GUI) screen 400A of an application that provides an assistant (e.g., a copilot) for utilizing the application. In an example, the assistant if provided for a communications application such as Microsoft Teams. As depicted, the assistant may begin the conversation (e.g., when a user invokes the assistance) by displaying a UI element 410 that asks how it can help the user. The user can utilize a UI element 412 to enter a user query. In the example depicted in screen 400A, the user seeks guidance in creating a calendar meeting. In response, the assistant (e.g., language model) provides summarized instructions on how to create a calendar meeting using the application by displaying the instructions in a UI element 414; [0045], FIG. 4B depicts an alternative example of displaying the response. The response depicted in the UI element 420 of FIG. 4B includes more detailed instructions on how to create a new meeting in the application; [0051], Once the prompt is transmitted to the model, a response to the product help inquiry is received from the model, at 510. The response may include a set of instructions (e.g., an ordered list of actions) to follow to achieve the results indicated in the product help inquiry. This list of actions may be referred to as a path of actions. Method 500 proceeds to extract this path of actions from the response, at 512. This may involve use of an ML model and/or one or more classifiers and utilizing a resource code for the application to identify commands provided by the application or UI elements used by the application that are included in the response) It would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to have incorporated the instructions indicating an interactive element selected from the set of interactive elements and prompting user interaction with the interactive element via the user interface in performance of the next action as suggested in Mathur into Fabian in view of Mathur. Doing so would be desirable because recently, there has been a significant increase in the use of machine-learning (ML) models such as large language models (LLM) to provide a variety of services and functions. Many LLMs receive an input such as a text segment and provide a prediction based on the input. Because of the wide use of LLMs, it is important that such models provide accurate results. However, even when aggregate accuracy is high, LLMs often fail when providing responses relating to specific domains such as specific products or topics (see Mathur [0001]). Hence, there is a need for improved systems and methods of evaluating responses provided by LLMs (see Mathur [0002]). A user can ask the chatbot how to perform a specific function within the application (e.g., how do I create a calendar event?). This provides an easy-to-use interface by which users can determine how to use an application or any other product. However, in order to ensure that the LLM provides accurate responses to user questions, the LLM would need finetuning with information related to the specific application or product. Finetuning an LLM, however, is often an expensive undertaking. That is because when focusing on a specific domain, it is often difficult to test and engineer prompts in a way that is scalable and has sufficient samples. Moreover, even a finetuned LLM may still exhibit failures in specific domains of data. For example, the LLM may provide harmful, inaccurate or irrelevant responses to specific queries. In an example, the response given might be written well but might have incomplete knowledge of the query. In another example, the response may miss certain steps that are crucial information for the user in performing an action. Such responses result in user frustration, inefficiency and erosion of user trust in the product. Thus, when considering integrations and implementations of LLMs in various tools and applications, it is important to have a reliable metric for evaluating the model's success and the specificity of the model to a specific domain (e.g., specific product). This is particularly true for an LLM that is finetuned to provide responses to product help inquiries, where the response may need to include a sequential number of steps for performing an action in the application (see Mathur [0015]). The system of Mathur can be used with a variety of applications, such as a communications application (e.g., Microsoft® Teams®), presentation application, design application, word processing application, spreadsheet application, social media application, and any other application for which a user may require help (see Mathur [0021]). Additionally, the system of Mathur would improve the system of Fabian by clarifying that the user may manually implement the steps provided in a response (see Fabian Figs. 7-10). Allowing the user to determine how to proceed based on the response would provide improved flexibility in situations where the user may prefer to perform an action themselves rather than rely on the system, which would increase ease of use and user satisfaction. Regarding claim 2, Fabian in view of Mathur teaches all the limitations of claim 1, further comprising: wherein generating, using the one or more large language models, the instructions to perform the next action in the sequence for performing the task comprises generating, using an additional large language model, an operation to execute in performing the next action (Fabian Figs 1-13; [0028], the reply from the LLM includes a specific suggestion for modifying the spreadsheet in the specific way; The LLM, upon receiving the configured prompt, generates a reply which suggests adding a column which computes a total sales quantity based on year-to-date (YTD) sales data along with a spreadsheet formula to calculate the quantity; the user may then direct the application to implement; [0048-0049], Upon receiving the user's selection of the first suggestion in task pane 144, application service 110 implements the suggestion by adding a column (not shown) to the spreadsheet data; [0084-0087], The user clicks the “Add column” button, causing one or more of application components 303 to modify data table 502 to include the new column and to fill the column according to instructions provided in the reply; [0088-0089], LLM 330 generates two formulas along with the descriptions and explanations and determines, in FIG. 5F, that the formulas are of moderate and high relevance to the input. LLM 330 transmits its reply to prompt engine 305 which, in turn, generates a response and sends the response to user interface 307 for display; [0094-0095], the application service inputs a second prompt to the LLM service which instructs the LLM service to provide a chain-of-thought breakdown or decomposition of the formula generated for the first prompt; [0097-0098], In FIG. 7B, upon receiving the formula in response to the first prompt, prompt engine 305 generates a second prompt which tasks LLM 330 with using Excel's LET function to break the formula down into steps; [0099-0101], FIGS. 8A, 8B, and 8C illustrate operational scenario 800 which presents an alternative presentation of the reply to the second prompt from LLM 330 beginning with, in FIG. 8A, receiving the user input requesting a way to calculate the surface area of a cylinder; Based on the reply from LLM 330 to the second prompt, prompt engine 305 configures and progressively displays the parts or steps of the output in user interface 307. As each part is displayed, the user can manually advance to the disclosure of the next part, for example, by clicking an “Accept.”; [0102-0103], In FIG. 9A, the user selects a cell next to the last column of the data table. The application floats a suggestion to allow the application to create a column for the user; [0107-0111], In FIG. 10C, task pane 1010 is updated by the application to show input 1016 and output 1017 based on the LLM's reply to the prompt based on input 1016. Output 1017 includes three suggestions received from the LLM in response to input 1016 according to parameters provided by the application in the prompt) Regarding claim 3, Fabian in view of Mathur teaches all the limitations of claim 2, further comprising: wherein generating, using the one or more large language models, the instructions to perform the next action in the sequence for performing the task comprises determining, using the large language model, to target the interactive element selected from the set of interactive elements via the operation to execute in performing the next action (Fabian Figs 1-13; [0028], the reply from the LLM includes a specific suggestion for modifying the spreadsheet in the specific way; The LLM, upon receiving the configured prompt, generates a reply which suggests adding a column which computes a total sales quantity based on year-to-date (YTD) sales data along with a spreadsheet formula to calculate the quantity; the user may then direct the application to implement; [0048-0049], Upon receiving the user's selection of the first suggestion in task pane 144, application service 110 implements the suggestion by adding a column (not shown) to the spreadsheet data; [0084-0087], The user clicks the “Add column” button, causing one or more of application components 303 to modify data table 502 to include the new column and to fill the column according to instructions provided in the reply; [0088-0089], LLM 330 generates two formulas along with the descriptions and explanations and determines, in FIG. 5F, that the formulas are of moderate and high relevance to the input. LLM 330 transmits its reply to prompt engine 305 which, in turn, generates a response and sends the response to user interface 307 for display; [0094-0095], the application service inputs a second prompt to the LLM service which instructs the LLM service to provide a chain-of-thought breakdown or decomposition of the formula generated for the first prompt; [0097-0098], In FIG. 7B, upon receiving the formula in response to the first prompt, prompt engine 305 generates a second prompt which tasks LLM 330 with using Excel's LET function to break the formula down into steps; [0099-0101], FIGS. 8A, 8B, and 8C illustrate operational scenario 800 which presents an alternative presentation of the reply to the second prompt from LLM 330 beginning with, in FIG. 8A, receiving the user input requesting a way to calculate the surface area of a cylinder; Based on the reply from LLM 330 to the second prompt, prompt engine 305 configures and progressively displays the parts or steps of the output in user interface 307. As each part is displayed, the user can manually advance to the disclosure of the next part, for example, by clicking an “Accept.”; [0102-0103], In FIG. 9A, the user selects a cell next to the last column of the data table. The application floats a suggestion to allow the application to create a column for the user; [0107-0111], In FIG. 10C, task pane 1010 is updated by the application to show input 1016 and output 1017 based on the LLM's reply to the prompt based on input 1016. Output 1017 includes three suggestions received from the LLM in response to input 1016 according to parameters provided by the application in the prompt) Regarding claim 4, Fabian in view of Mathur teaches all the limitations of claim 3, further comprising: further comprising generating a target element prompt that includes chain-of-thought reasoning, the chain-of-thought reasoning related to selection of a target interactive element based on a determined operation; wherein determining, using the large language model, to target the interactive element via the operation comprises determining, to target the interactive element by incorporating the chain-of-thought reasoning of the target element prompt via the large language model (Fabian Figs 1-13; [0028], the reply from the LLM includes a specific suggestion for modifying the spreadsheet in the specific way; The LLM, upon receiving the configured prompt, generates a reply which suggests adding a column which computes a total sales quantity based on year-to-date (YTD) sales data along with a spreadsheet formula to calculate the quantity; the user may then direct the application to implement; [0048-0049], Upon receiving the user's selection of the first suggestion in task pane 144, application service 110 implements the suggestion by adding a column (not shown) to the spreadsheet data; [0084-0087], The user clicks the “Add column” button, causing one or more of application components 303 to modify data table 502 to include the new column and to fill the column according to instructions provided in the reply; [0088-0089], LLM 330 generates two formulas along with the descriptions and explanations and determines, in FIG. 5F, that the formulas are of moderate and high relevance to the input. LLM 330 transmits its reply to prompt engine 305 which, in turn, generates a response and sends the response to user interface 307 for display; [0094], the application service inputs a second prompt to the LLM service which instructs the LLM service to provide a chain-of-thought breakdown or decomposition of the formula generated for the first prompt; [0095], In an implementation of a chain-of-thought breakdown, the second prompt generated by the application service directs the LLM service to return the formula generated in response to the first prompt in terms of the Microsoft Excel LET function; [0096], The second formula, in the same programming language as the first formula, is a chain-of-thought decomposition of the first formula into multiple parts such that executing the multiple parts in sequence effects the operation of the first formula; [0097], FIGS. 7A and 7B illustrate operational scenario 700 of prompting an LLM service by an application service involving the use of chain-of-thought decomposition; In FIG. 7B, upon receiving the formula in response to the first prompt, prompt engine 305 generates a second prompt which tasks LLM 330 with using Excel's LET function to break the formula down into steps; [0099-0101], FIGS. 8A, 8B, and 8C illustrate operational scenario 800 which presents an alternative presentation of the reply to the second prompt from LLM 330 beginning with, in FIG. 8A, receiving the user input requesting a way to calculate the surface area of a cylinder; Based on the reply from LLM 330 to the second prompt, prompt engine 305 configures and progressively displays the parts or steps of the output in user interface 307. As each part is displayed, the user can manually advance to the disclosure of the next part, for example, by clicking an “Accept.”; [0102-0103], In FIG. 9A, the user selects a cell next to the last column of the data table. The application floats a suggestion to allow the application to create a column for the user; [0107-0111], In FIG. 10C, task pane 1010 is updated by the application to show input 1016 and output 1017 based on the LLM's reply to the prompt based on input 1016. Output 1017 includes three suggestions received from the LLM in response to input 1016 according to parameters provided by the application in the prompt) Regarding claim 7, Fabian in view of Mathur teaches all the limitations of claim 1, further comprising: further comprising determining a set of previous actions that have been executed in performance of the task, wherein generating the estimated lookahead plan describing the one or more actions for performing the task comprises generating the estimated lookahead plan describing the one or more actions to be executed after the set of previous actions in performing the task (Fabian Figs 1-13; [0028], the reply from the LLM includes a specific suggestion for modifying the spreadsheet in the specific way; The LLM, upon receiving the configured prompt, generates a reply which suggests adding a column which computes a total sales quantity based on year-to-date (YTD) sales data along with a spreadsheet formula to calculate the quantity; the user may then direct the application to implement; [0032], To generate a prompt based on a user's natural language input, the application will configure the prompt according to one or more of: the scope of the problem, the tasks to be completed, illustrative examples, sample data or spreadsheet contextual information, rules, the output format, and a cue to get the LLM to complete the task and not ramble; [0048-0049], Upon receiving the user's selection of the first suggestion in task pane 144, application service 110 implements the suggestion by adding a column (not shown) to the spreadsheet data; [0060], Application service 110 receives the input and generates a prompt to be submitted to LLM service 120 based on the input. The prompt includes contextual information, such as a chat history and spreadsheet data or metadata, such as a table name, worksheet name, or spreadsheet name, row and column headers, and at least a subset of the data, e.g., the first five or ten rows of data. Contextual information can also include the recent or latest actions performed by the user on the spreadsheet, such as the user creating a new column or entering a formula; [0084-0087], The user clicks the “Add column” button, causing one or more of application components 303 to modify data table 502 to include the new column and to fill the column according to instructions provided in the reply; [0088-0089], LLM 330 generates two formulas along with the descriptions and explanations and determines, in FIG. 5F, that the formulas are of moderate and high relevance to the input. LLM 330 transmits its reply to prompt engine 305 which, in turn, generates a response and sends the response to user interface 307 for display; [0094-0095], the application service inputs a second prompt to the LLM service which instructs the LLM service to provide a chain-of-thought breakdown or decomposition of the formula generated for the first prompt; [0097-0098], In FIG. 7B, upon receiving the formula in response to the first prompt, prompt engine 305 generates a second prompt which tasks LLM 330 with using Excel's LET function to break the formula down into steps; [0099-0101], FIGS. 8A, 8B, and 8C illustrate operational scenario 800 which presents an alternative presentation of the reply to the second prompt from LLM 330 beginning with, in FIG. 8A, receiving the user input requesting a way to calculate the surface area of a cylinder; Based on the reply from LLM 330 to the second prompt, prompt engine 305 configures and progressively displays the parts or steps of the output in user interface 307. As each part is displayed, the user can manually advance to the disclosure of the next part, for example, by clicking an “Accept.”; [0102-0103], In FIG. 9A, the user selects a cell next to the last column of the data table. The application floats a suggestion to allow the application to create a column for the user; [0107-0111], In FIG. 10C, task pane 1010 is updated by the application to show input 1016 and output 1017 based on the LLM's reply to the prompt based on input 1016. Output 1017 includes three suggestions received from the LLM in response to input 1016 according to parameters provided by the application in the prompt) Regarding claim 8, Fabian in view of Mathur teaches all the limitations of claim 1, further comprising: further comprising determining an environment representation of the user interface, wherein generating the lookahead prompt comprising the at least one execution example corresponding to the task comprises generating the lookahead prompt comprising the at least one execution example and the environment representation (Fabian Figs 1-13; [0028], the reply from the LLM includes a specific suggestion for modifying the spreadsheet in the specific way; The LLM, upon receiving the configured prompt, generates a reply which suggests adding a column which computes a total sales quantity based on year-to-date (YTD) sales data along with a spreadsheet formula to calculate the quantity; the user may then direct the application to implement; [0032], To generate a prompt based on a user's natural language input, the application will configure the prompt according to one or more of: the scope of the problem, the tasks to be completed, illustrative examples, sample data or spreadsheet contextual information, rules, the output format, and a cue to get the LLM to complete the task and not ramble. The scope of the problem refers to the domain in which the problem is to be addressed. For example, the prompt may instruct the LLM to limit its reply to Microsoft Excel® formulas and not Google Sheets formulas. The tasks to be completed include instructions, such as, “generate a formula to do X and self-evaluate the formula.” Illustrative examples can include an example of how to self-evaluate a formula when the formula is relatively complex. An illustrative example provides the LLM with guidance on how to complete a task of the prompt; [0043], The spreadsheet environment of application service 110 may be implemented a natively installed and executed application, a browser-based application, or a mobile application; [0048-0049], Upon receiving the user's selection of the first suggestion in task pane 144, application service 110 implements the suggestion by adding a column (not shown) to the spreadsheet data; [0060], Application service 110 receives the input and generates a prompt to be submitted to LLM service 120 based on the input. The prompt includes contextual information, such as a chat history and spreadsheet data or metadata, such as a table name, worksheet name, or spreadsheet name, row and column headers, and at least a subset of the data, e.g., the first five or ten rows of data. Contextual information can also include the recent or latest actions performed by the user on the spreadsheet, such as the user creating a new column or entering a formula; [0084-0087], The user clicks the “Add column” button, causing one or more of application components 303 to modify data table 502 to include the new column and to fill the column according to instructions provided in the reply; [0088-0089], LLM 330 generates two formulas along with the descriptions and explanations and determines, in FIG. 5F, that the formulas are of moderate and high relevance to the input. LLM 330 transmits its reply to prompt engine 305 which, in turn, generates a response and sends the response to user interface 307 for display; [0094-0095], the application service inputs a second prompt to the LLM service which instructs the LLM service to provide a chain-of-thought breakdown or decomposition of the formula generated for the first prompt; [0097-0098], In FIG. 7B, upon receiving the formula in response to the first prompt, prompt engine 305 generates a second prompt which tasks LLM 330 with using Excel's LET function to break the formula down into steps; [0099-0101], FIGS. 8A, 8B, and 8C illustrate operational scenario 800 which presents an alternative presentation of the reply to the second prompt from LLM 330 beginning with, in FIG. 8A, receiving the user input requesting a way to calculate the surface area of a cylinder; Based on the reply from LLM 330 to the second prompt, prompt engine 305 configures and progressively displays the parts or steps of the output in user interface 307. As each part is displayed, the user can manually advance to the disclosure of the next part, for example, by clicking an “Accept.”; [0102-0103], In FIG. 9A, the user selects a cell next to the last column of the data table. The application floats a suggestion to allow the application to create a column for the user; [0107-0111], In FIG. 10C, task pane 1010 is updated by the application to show input 1016 and output 1017 based on the LLM's reply to the prompt based on input 1016. Output 1017 includes three suggestions received from the LLM in response to input 1016 according to parameters provided by the application in the prompt) Regarding claim 10, Fabian teaches the claim comprising: A system comprising: one or more memory devices; and one or more processors configured to cause the system to (Fabian Figs. 1-13; [0119], FIG. 13 illustrates computing device 1301 that is representative of any system or collection of systems in which the various processes, programs, services, and scenarios disclosed herein may be implemented. Examples of computing device 1301 include, but are not limited to, desktop and laptop computers, tablet computers, mobile computers, and wearable devices; [0122], processing system 1302 may comprise a micro-processor and other circuitry that retrieves and executes software 1305 from storage system 130) : receive a query regarding a task to be performed via a user interface of a software application, the software application having a set of interactive elements that are accessible via the user interface; determine, from the set of interactive elements of the software application, a set of candidate interactive elements that correspond to performance of the task (Fabian Figs 1-13; [0032], To generate a prompt based on a user's natural language input, the application will configure the prompt according to one or more of: the scope of the problem, the tasks to be completed, illustrative examples, sample data or spreadsheet contextual information, rules, the output format, and a cue to get the LLM to complete the task and not ramble. The scope of the problem refers to the domain in which the problem is to be addressed. For example, the prompt may instruct the LLM to limit its reply to Microsoft Excel® formulas and not Google Sheets formulas. The tasks to be completed include instructions, such as, “generate a formula to do X and self-evaluate the formula.” Illustrative examples can include an example of how to self-evaluate a formula when the formula is relatively complex. An illustrative example provides the LLM with guidance on how to complete a task of the prompt; [0060], Application service 110 receives the input and generates a prompt to be submitted to LLM service 120 based on the input. The prompt includes contextual information, such as a chat history and spreadsheet data or metadata, such as a table name, worksheet name, or spreadsheet name, row and column headers, and at least a subset of the data, e.g., the first five or ten rows of data. Contextual information can also include the recent or latest actions performed by the user on the spreadsheet, such as the user creating a new column or entering a formula. Contextual information can also include errors detected in the spreadsheet; [0066], To configure the prompt, prompt engine 305 identifies a prompt template according to the type of request in the input in an implementation. Prompt templates can include prompt configurations for suggesting a calculated column to be added to workbook data 320, for a general inquiry about workbook data 320, for analyzing data in workbook data 320 to project a result, such as for a hypothetical scenario, and so on; [0067], Prompt engine 305 configures the prompt including parameters to direct LLM 330 to provide a focused response to the input. Prompt parameters include the scope of the prompt, the output format of the reply to produce a reply in a parse-able format, instructions or tasks, examples including sample data or sample data formatting, special tokens which influence the behavior of LLM 330, and so on; [0082], FIGS. 5B-5F continue operational scenario 500, illustrating a turn-based chat including inputs received from a user, prompts generated based on the inputs, and responses based on replies to the inputs from LLM 330. In FIG. 5B, the user enters the natural language inquiry about data table 502 into task pane 503. User interface 307 transmits the user input to prompt engine 305 which generates a prompt based on the input. Prompt engine 305 selects a prompt template based on a type or classification of the inquiry and configures a prompt according to the template. In the prompt, prompt engine 305 includes tasks, instructions, or rules applicable to generating a reply to the input, such as tasking LLM 330 to perform a self-evaluation of its reply and a rule to return the reply in a particular output format which is suitable for parsing; [0083], Prompt engine 305 also includes contextual information in the prompt. Prompt engine 305 retrieves spreadsheet context data relating to data table 502 from various ones of application components 303. For example, for a general inquiry about the contents of data table 502, prompt engine 305 configures the prompt to include all the column headers in data table 502 along with the table name and the first five rows of data; [0099-0101], FIGS. 8A, 8B, and 8C illustrate operational scenario 800 which presents an alternative presentation of the reply to the second prompt from LLM 330 beginning with, in FIG. 8A, receiving the user input requesting a way to calculate the surface area of a cylinder; [0108], FIGS. 10A, 10B, and 10C illustrate operational scenario 1000 of an LLM integration with a spreadsheet environment) ; generate, from the set of candidate interactive elements and using a first large language model, an estimated lookahead plan describing one or more actions for performing the task, the one or more actions involving use of one or more interactive elements from the set of interactive elements; generate, from the set of candidate interactive elements and using a second large language model, an operation for a next action in a sequence for performing the task; determine, using the first large language model and from the set of candidate interactive elements and from the estimated lookahead plan, an interactive element from the set of candidate interactive elements to target via the operation of the next action; and generate instructions that indicate the interactive element (Fabian Figs 1-13; [0028], the reply from the LLM includes a specific suggestion for modifying the spreadsheet in the specific way; The LLM, upon receiving the configured prompt, generates a reply which suggests adding a column which computes a total sales quantity based on year-to-date (YTD) sales data along with a spreadsheet formula to calculate the quantity. Based on the reply, the application displays the suggestion along with the formula in the user interface of the spreadsheet. In an implementation, the application generates a preview of the column to be added to the data table based on the suggestion, i.e., a column calculating YTD sales data based on the table data and using the formula provided in the reply of the LLM, which the user may then direct the application to implement; [0048], Application service 110 may display the suggestions according to how LLM service 120 self-evaluated the suggestions, e.g., according to relevance to the input or correctness; [0049], Upon receiving the user's selection of the first suggestion in task pane 144, application service 110 implements the suggestion by adding a column (not shown) to the spreadsheet data; [0084], Continuing with FIG. 5B, with the prompt configured, prompt engine 305 sends the prompt to LLM 330. LLM 330 generates a reply to the prompt and transmits the reply to prompt engine 305; [0087], Upon receiving LLM 330's reply, prompt engine 305 post-processes the reply to generate a response for display in user interface 307. Post-processing the reply includes extracting information from the reply according to the output formatting rules, such as a description of the suggestion and instructions by which the suggestion can be implemented either by application components 303 or by the user. Continuing to FIG. 5D, the response includes a link by which the user can view more detailed information about the suggestion and a graphical button by which to add the column to data table 502. The user clicks the “Add column” button, causing one or more of application components 303 to modify data table 502 to include the new column and to fill the column according to instructions provided in the reply; [0088-0089], LLM 330 generates two formulas along with the descriptions and explanations and determines, in FIG. 5F, that the formulas are of moderate and high relevance to the input. LLM 330 transmits its reply to prompt engine 305 which, in turn, generates a response and sends the response to user interface 307 for display; [0094], the application service inputs a second prompt to the LLM service which instructs the LLM service to provide a chain-of-thought breakdown or decomposition of the formula generated for the first prompt; [0095], the second prompt generated by the application service directs the LLM service to return the formula generated in response to the first prompt in terms of the Microsoft Excel LET function; [0097]. FIGS. 7A and 7B illustrate operational scenario 700 of prompting an LLM service by an application service involving the use of chain-of-thought decomposition; [0098], In FIG. 7B, upon receiving the formula in response to the first prompt, prompt engine 305 generates a second prompt which tasks LLM 330 with using Excel's LET function to break the formula down into steps; [0099-0101], FIGS. 8A, 8B, and 8C illustrate operational scenario 800 which presents an alternative presentation of the reply to the second prompt from LLM 330 beginning with, in FIG. 8A, receiving the user input requesting a way to calculate the surface area of a cylinder; Based on the reply from LLM 330 to the second prompt, prompt engine 305 configures and progressively displays the parts or steps of the output in user interface 307. As each part is displayed, the user can manually advance to the disclosure of the next part, for example, by clicking an “Accept.”; [0102], FIGS. 9A-9E illustrate operational scenario 900; [0103], In FIG. 9A, the user selects a cell next to the last column of the data table. The application floats a suggestion to allow the application to create a column for the user; [0107], FIG. 9D illustrates task pane 910 subsequent to the user selecting the Years of Service column for inclusion in the data table. In textbox 911, the user enters a natural language input to “Add a column with the employee's last name.” The application configures a prompt based on the input which tasks the LLM to produce suggestions in response to the input; [0108], FIGS. 10A, 10B, and 10C illustrate operational scenario 1000 of an LLM integration with a spreadsheet environment; [0109], The user is also presented with textbox 1013 by which the user can submit natural language inputs, such as requests or queries; [0110], The LLM generates and sends a reply to the prompt to the application which the application processes to produce a response to the input. The response to the input from the application based on the reply from the LLM is shown in output 1015. Output 1015 includes an explanation of the reply and a table demonstrating a spreadsheet calculation, including spreadsheet formulas, demonstrating a calculation responsive to the input; [0111], In FIG. 10C, task pane 1010 is updated by the application to show input 1016 and output 1017 based on the LLM's reply to the prompt based on input 1016. Output 1017 includes three suggestions received from the LLM in response to input 1016 according to parameters provided by the application in the prompt) However, Fabian fails to expressly disclose generate instructions that indicate the interactive element and prompt user interaction with the interactive element via the user interface in execution of the operation. In the same field of endeavor, Mathur teaches: generate instructions that indicate the interactive element and prompt user interaction with the interactive element via the user interface in execution of the operation (Mathur Figs. 1-7; [0015], the response may miss certain steps that are crucial information for the user in performing an action; the response may need to include a sequential number of steps for performing an action in the application; [0021], The application 120 and local applications 164A-164N (collectively referred to as local application 164) may be any application that enables a user such as users 162A-162N (collectively referred to as user 162) to interact with the application to perform an action or achieve a purpose; [0024], the application 120 and/or application 164 utilize the LLM 130 to provide an application copilot that assists users in navigating application features and/or performing actions within the application; [0036], The path extraction engine 240 examines the response to extract a relevant path of actions included in the response. The terms “action path”, “path of actions” or path as used herein refer to a sequence of terms included in a response from the language model or in a help documentation, where each term refers to an action a user takes or refers to a user interface elements a user selects to take an action to perform a task associated with the product help inquiry., the extracted action path is a sequence of terms used in the response that correspond with specific actions or user interface elements in the product. For example, the action path for the example response depicted in FIG. 4B which explains how to create a new calendar meeting, is “New Meeting”, “Invite People”, and “Send.” The actions correspond with specific user interface elements of the meeting application. The path extraction engine 240 may utilize information about the product/application to retrieve key terms in the response that are associated with product actions/UI elements; [0044], FIGS. 4A-4C depict example user interface (UI) screens of an application that utilizes a language model to provide responses to product help inquiries. FIG. 4A depicts an example graphical user interface (GUI) screen 400A of an application that provides an assistant (e.g., a copilot) for utilizing the application. In an example, the assistant if provided for a communications application such as Microsoft Teams. As depicted, the assistant may begin the conversation (e.g., when a user invokes the assistance) by displaying a UI element 410 that asks how it can help the user. The user can utilize a UI element 412 to enter a user query. In the example depicted in screen 400A, the user seeks guidance in creating a calendar meeting. In response, the assistant (e.g., language model) provides summarized instructions on how to create a calendar meeting using the application by displaying the instructions in a UI element 414; [0045], FIG. 4B depicts an alternative example of displaying the response. The response depicted in the UI element 420 of FIG. 4B includes more detailed instructions on how to create a new meeting in the application; [0051], Once the prompt is transmitted to the model, a response to the product help inquiry is received from the model, at 510. The response may include a set of instructions (e.g., an ordered list of actions) to follow to achieve the results indicated in the product help inquiry. This list of actions may be referred to as a path of actions. Method 500 proceeds to extract this path of actions from the response, at 512. This may involve use of an ML model and/or one or more classifiers and utilizing a resource code for the application to identify commands provided by the application or UI elements used by the application that are included in the response) It would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to have incorporated generate instructions that indicate the interactive element and prompt user interaction with the interactive element via the user interface in execution of the operation as suggested in Mathur into Fabian in view of Mathur. Doing so would be desirable because Recently, there has been a significant increase in the use of machine-learning (ML) models such as large language models (LLM) to provide a variety of services and functions. Many LLMs receive an input such as a text segment and provide a prediction based on the input. Because of the wide use of LLMs, it is important that such models provide accurate results. However, even when aggregate accuracy is high, LLMs often fail when providing responses relating to specific domains such as specific products or topics (see Mathur [0001]). Hence, there is a need for improved systems and methods of evaluating responses provided by LLMs (see Mathur [0002]). A user can ask the chatbot how to perform a specific function within the application (e.g., how do I create a calendar event?). This provides an easy-to-use interface by which users can determine how to use an application or any other product. However, in order to ensure that the LLM provides accurate responses to user questions, the LLM would need finetuning with information related to the specific application or product. Finetuning an LLM, however, is often an expensive undertaking. That is because when focusing on a specific domain, it is often difficult to test and engineer prompts in a way that is scalable and has sufficient samples. Moreover, even a finetuned LLM may still exhibit failures in specific domains of data. For example, the LLM may provide harmful, inaccurate or irrelevant responses to specific queries. In an example, the response given might be written well but might have incomplete knowledge of the query. In another example, the response may miss certain steps that are crucial information for the user in performing an action. Such responses result in user frustration, inefficiency and erosion of user trust in the product. Thus, when considering integrations and implementations of LLMs in various tools and applications, it is important to have a reliable metric for evaluating the model's success and the specificity of the model to a specific domain (e.g., specific product). This is particularly true for an LLM that is finetuned to provide responses to product help inquiries, where the response may need to include a sequential number of steps for performing an action in the application (see Mathur [0015]). The system of Mathur can be used with a variety of applications, such as a communications application (e.g., Microsoft® Teams®), presentation application, design application, word processing application, spreadsheet application, social media application, and any other application for which a user may require help (see Mathur [0021]). Additionally, the system of Mathur would improve the system of Fabian by clarifying that the user may manually implement the steps provided in a response (see Fabian Figs. 7-10). Allowing the user to determine how to proceed based on the response would provide improved flexibility in situations where the user may prefer to perform an action themselves rather than rely on the system, which would increase ease of use and user satisfaction. Regarding claim 11, Fabian in view of Mathur teaches all the limitations of claim 10, further comprising: wherein the one or more processors are configured to cause the system to determine the set of candidate interactive elements of the user interface by: extracting, from an environment representation of the user interface, a plurality of interactive elements; generating, using a ranking model and from the query and from one or more previous actions that have been executed in performance of the task, a ranking of the plurality of interactive elements; and determining the set of candidate interactive elements from the plurality of interactive elements based on the ranking (Fabian Figs 1-13; [0027], the user's general inquiry is ambiguous or underspecified, and the application prompts the LLM to interpret the reply in multiple ways and to generate suggestions based on the multiple interpretations; The application may include in its prompt to the LLM an instruction to interpret the inquiry in multiple ways and to generate suggestions based on the interpretations; [0028], the reply from the LLM includes a specific suggestion for modifying the spreadsheet in the specific way; The LLM, upon receiving the configured prompt, generates a reply which suggests adding a column which computes a total sales quantity based on year-to-date (YTD) sales data along with a spreadsheet formula to calculate the quantity; the user may then direct the application to implement; [0032], To generate a prompt based on a user's natural language input, the application will configure the prompt according to one or more of: the scope of the problem, the tasks to be completed, illustrative examples, sample data or spreadsheet contextual information, rules, the output format, and a cue to get the LLM to complete the task and not ramble. The scope of the problem refers to the domain in which the problem is to be addressed. For example, the prompt may instruct the LLM to limit its reply to Microsoft Excel® formulas and not Google Sheets formulas. The tasks to be completed include instructions, such as, “generate a formula to do X and self-evaluate the formula.” Illustrative examples can include an example of how to self-evaluate a formula when the formula is relatively complex; [0036], The application may order the alternative suggestions or eliminate suggestions based on their self-evaluations and present the alternative suggestions in the task pane of the user interface based on the self-evaluations; [0037], if the user request is determined to be ambiguous, the application may direct the LLM to produce a formula and suggest alternative interpretations of the query; [0048-0049], suggested actions may be based on a higher-ranked suggestion received from LLM service 120 in response to the user's original input; [0084-0087], the user submits a second input in task pane 503 requesting suggestions for making “it” better. Prompt engine 305 selects a template to configure a prompt. For an ambiguous or underspecified input, the prompt template may include an instruction to LLM 330 to interpret the input in multiple ways and to generate multiple alternative suggestions based on the interpretations. Here, “it” may refer to, for example, the formatting of data table 502, the organization of data table 502, to the contents of data table 502, or to some other aspect of data table 502. Prompt engine 305 adds context information to the prompt: relevant spreadsheet data, such as the column headers and some rows of data, and the chat history thus far. The prompt also includes an instruction for LLM 330 to self-evaluate its suggestions and to filter out suggestions which are of less than moderate relevance to the subject of the inquiry. The tasks may also include a rule to generate, for any suggested formulas produced by LLM 330, a description and/or explanation of the formula enclosed in the appropriate tags; [0088-0089], LLM 330 generates two formulas along with the descriptions and explanations and determines, in FIG. 5F, that the formulas are of moderate and high relevance to the input. LLM 330 transmits its reply to prompt engine 305 which, in turn, generates a response and sends the response to user interface 307 for display; [0094], the application service inputs a second prompt to the LLM service which instructs the LLM service to provide a chain-of-thought breakdown or decomposition of the formula generated for the first prompt; [0095], In an implementation of a chain-of-thought breakdown, the second prompt generated by the application service directs the LLM service to return the formula generated in response to the first prompt in terms of the Microsoft Excel LET function; [0096], The second formula, in the same programming language as the first formula, is a chain-of-thought decomposition of the first formula into multiple parts such that executing the multiple parts in sequence effects the operation of the first formula; [0097], FIGS. 7A and 7B illustrate operational scenario 700 of prompting an LLM service by an application service involving the use of chain-of-thought decomposition; In FIG. 7B, upon receiving the formula in response to the first prompt, prompt engine 305 generates a second prompt which tasks LLM 330 with using Excel's LET function to break the formula down into steps; [0099-0101], FIGS. 8A, 8B, and 8C illustrate operational scenario 800 which presents an alternative presentation of the reply to the second prompt from LLM 330 beginning with, in FIG. 8A, receiving the user input requesting a way to calculate the surface area of a cylinder; Based on the reply from LLM 330 to the second prompt, prompt engine 305 configures and progressively displays the parts or steps of the output in user interface 307. As each part is displayed, the user can manually advance to the disclosure of the next part, for example, by clicking an “Accept.”; [0102-0103], In FIG. 9A, the user selects a cell next to the last column of the data table. The application floats a suggestion to allow the application to create a column for the user; [0107-0111], In FIG. 10C, task pane 1010 is updated by the application to show input 1016 and output 1017 based on the LLM's reply to the prompt based on input 1016. Output 1017 includes three suggestions received from the LLM in response to input 1016 according to parameters provided by the application in the prompt) Regarding claim 15, Fabian in view of Mathur teaches all the limitations of claim 10, further comprising: wherein the one or more processors are configured to cause the system to determine the interactive element from the set of candidate interactive elements and the estimated lookahead plan using the first large language model by: determining a set of target element examples, each target element example including an example task and an example interactive element, the example interactive element to be targeted by an example operation in performance of an example next action for the example task; and determining, using the first large language model and from the set of candidate interactive elements, the interactive element based on the estimated lookahead plan and the set of target element examples (Fabian Figs 1-13; [0027], the user's general inquiry is ambiguous or underspecified, and the application prompts the LLM to interpret the reply in multiple ways and to generate suggestions based on the multiple interpretations; The application may include in its prompt to the LLM an instruction to interpret the inquiry in multiple ways and to generate suggestions based on the interpretations; [0028], the reply from the LLM includes a specific suggestion for modifying the spreadsheet in the specific way; The LLM, upon receiving the configured prompt, generates a reply which suggests adding a column which computes a total sales quantity based on year-to-date (YTD) sales data along with a spreadsheet formula to calculate the quantity; the user may then direct the application to implement; [0032], To generate a prompt based on a user's natural language input, the application will configure the prompt according to one or more of: the scope of the problem, the tasks to be completed, illustrative examples, sample data or spreadsheet contextual information, rules, the output format, and a cue to get the LLM to complete the task and not ramble. The scope of the problem refers to the domain in which the problem is to be addressed. For example, the prompt may instruct the LLM to limit its reply to Microsoft Excel® formulas and not Google Sheets formulas. The tasks to be completed include instructions, such as, “generate a formula to do X and self-evaluate the formula.” Illustrative examples can include an example of how to self-evaluate a formula when the formula is relatively complex; [0036], The application may order the alternative suggestions or eliminate suggestions based on their self-evaluations and present the alternative suggestions in the task pane of the user interface based on the self-evaluations; [0037], if the user request is determined to be ambiguous, the application may direct the LLM to produce a formula and suggest alternative interpretations of the query; [0048-0049], suggested actions may be based on a higher-ranked suggestion received from LLM service 120 in response to the user's original input; [0084-0087], the user submits a second input in task pane 503 requesting suggestions for making “it” better. Prompt engine 305 selects a template to configure a prompt. For an ambiguous or underspecified input, the prompt template may include an instruction to LLM 330 to interpret the input in multiple ways and to generate multiple alternative suggestions based on the interpretations. Here, “it” may refer to, for example, the formatting of data table 502, the organization of data table 502, to the contents of data table 502, or to some other aspect of data table 502. Prompt engine 305 adds context information to the prompt: relevant spreadsheet data, such as the column headers and some rows of data, and the chat history thus far. The prompt also includes an instruction for LLM 330 to self-evaluate its suggestions and to filter out suggestions which are of less than moderate relevance to the subject of the inquiry. The tasks may also include a rule to generate, for any suggested formulas produced by LLM 330, a description and/or explanation of the formula enclosed in the appropriate tags; [0088-0089], LLM 330 generates two formulas along with the descriptions and explanations and determines, in FIG. 5F, that the formulas are of moderate and high relevance to the input. LLM 330 transmits its reply to prompt engine 305 which, in turn, generates a response and sends the response to user interface 307 for display; [0094], the application service inputs a second prompt to the LLM service which instructs the LLM service to provide a chain-of-thought breakdown or decomposition of the formula generated for the first prompt; [0095], In an implementation of a chain-of-thought breakdown, the second prompt generated by the application service directs the LLM service to return the formula generated in response to the first prompt in terms of the Microsoft Excel LET function; [0096], The second formula, in the same programming language as the first formula, is a chain-of-thought decomposition of the first formula into multiple parts such that executing the multiple parts in sequence effects the operation of the first formula; [0097], FIGS. 7A and 7B illustrate operational scenario 700 of prompting an LLM service by an application service involving the use of chain-of-thought decomposition; In FIG. 7B, upon receiving the formula in response to the first prompt, prompt engine 305 generates a second prompt which tasks LLM 330 with using Excel's LET function to break the formula down into steps; [0099-0101], FIGS. 8A, 8B, and 8C illustrate operational scenario 800 which presents an alternative presentation of the reply to the second prompt from LLM 330 beginning with, in FIG. 8A, receiving the user input requesting a way to calculate the surface area of a cylinder; Based on the reply from LLM 330 to the second prompt, prompt engine 305 configures and progressively displays the parts or steps of the output in user interface 307. As each part is displayed, the user can manually advance to the disclosure of the next part, for example, by clicking an “Accept.”; [0102-0103], In FIG. 9A, the user selects a cell next to the last column of the data table. The application floats a suggestion to allow the application to create a column for the user; [0107-0111], In FIG. 10C, task pane 1010 is updated by the application to show input 1016 and output 1017 based on the LLM's reply to the prompt based on input 1016. Output 1017 includes three suggestions received from the LLM in response to input 1016 according to parameters provided by the application in the prompt) Regarding claim 16, Fabian in view of Mathur teaches all the limitations of claim 15, further comprising: wherein determining, using the first large language model and from the set of candidate interactive elements, the interactive element based on the estimated lookahead plan and the set of target element examples comprises: generating a target element prompt that includes chain-of-thought reasoning, the chain-of-thought reasoning related to selection of a target interactive element based on a determined operation, the chain-of-thought reasoning incorporating the set of candidate interactive elements, the estimated lookahead plan, and the set of target element examples; and determining, using the first large language model and based on the target element prompt, the interactive element (Fabian Figs 1-13; [0028], the reply from the LLM includes a specific suggestion for modifying the spreadsheet in the specific way; The LLM, upon receiving the configured prompt, generates a reply which suggests adding a column which computes a total sales quantity based on year-to-date (YTD) sales data along with a spreadsheet formula to calculate the quantity; the user may then direct the application to implement; [0048-0049], Upon receiving the user's selection of the first suggestion in task pane 144, application service 110 implements the suggestion by adding a column (not shown) to the spreadsheet data; [0084-0087], The user clicks the “Add column” button, causing one or more of application components 303 to modify data table 502 to include the new column and to fill the column according to instructions provided in the reply; [0088-0089], LLM 330 generates two formulas along with the descriptions and explanations and determines, in FIG. 5F, that the formulas are of moderate and high relevance to the input. LLM 330 transmits its reply to prompt engine 305 which, in turn, generates a response and sends the response to user interface 307 for display; [0094], the application service inputs a second prompt to the LLM service which instructs the LLM service to provide a chain-of-thought breakdown or decomposition of the formula generated for the first prompt; [0095], In an implementation of a chain-of-thought breakdown, the second prompt generated by the application service directs the LLM service to return the formula generated in response to the first prompt in terms of the Microsoft Excel LET function; [0096], The second formula, in the same programming language as the first formula, is a chain-of-thought decomposition of the first formula into multiple parts such that executing the multiple parts in sequence effects the operation of the first formula; [0097], FIGS. 7A and 7B illustrate operational scenario 700 of prompting an LLM service by an application service involving the use of chain-of-thought decomposition; In FIG. 7B, upon receiving the formula in response to the first prompt, prompt engine 305 generates a second prompt which tasks LLM 330 with using Excel's LET function to break the formula down into steps; [0099-0101], FIGS. 8A, 8B, and 8C illustrate operational scenario 800 which presents an alternative presentation of the reply to the second prompt from LLM 330 beginning with, in FIG. 8A, receiving the user input requesting a way to calculate the surface area of a cylinder; Based on the reply from LLM 330 to the second prompt, prompt engine 305 configures and progressively displays the parts or steps of the output in user interface 307. As each part is displayed, the user can manually advance to the disclosure of the next part, for example, by clicking an “Accept.”; [0102-0103], In FIG. 9A, the user selects a cell next to the last column of the data table. The application floats a suggestion to allow the application to create a column for the user; [0107-0111], In FIG. 10C, task pane 1010 is updated by the application to show input 1016 and output 1017 based on the LLM's reply to the prompt based on input 1016. Output 1017 includes three suggestions received from the LLM in response to input 1016 according to parameters provided by the application in the prompt) Regarding claim 17, Fabian in view of Mathur teaches all the limitations of claim 10, further comprising: wherein the one or more processors are further configured to cause the system to: generate the operation for the next action in the sequence for performing the task by determining that the operation includes a type operation or a select operation; and determine a value as an argument for the type operation or the select operation (Fabian Figs 1-13; [0028], the reply from the LLM includes a specific suggestion for modifying the spreadsheet in the specific way; The LLM, upon receiving the configured prompt, generates a reply which suggests adding a column which computes a total sales quantity based on year-to-date (YTD) sales data along with a spreadsheet formula to calculate the quantity; the user may then direct the application to implement; [0032], To generate a prompt based on a user's natural language input, the application will configure the prompt according to one or more of: the scope of the problem, the tasks to be completed, illustrative examples, sample data or spreadsheet contextual information, rules, the output format, and a cue to get the LLM to complete the task and not ramble. The scope of the problem refers to the domain in which the problem is to be addressed. For example, the prompt may instruct the LLM to limit its reply to Microsoft Excel® formulas and not Google Sheets formulas. The tasks to be completed include instructions, such as, “generate a formula to do X and self-evaluate the formula.”; [0048-0049], Upon receiving the user's selection of the first suggestion in task pane 144, application service 110 implements the suggestion by adding a column (not shown) to the spreadsheet data; [0084-0087], The user clicks the “Add column” button, causing one or more of application components 303 to modify data table 502 to include the new column and to fill the column according to instructions provided in the reply; [0088-0089], LLM 330 generates two formulas along with the descriptions and explanations and determines, in FIG. 5F, that the formulas are of moderate and high relevance to the input. LLM 330 transmits its reply to prompt engine 305 which, in turn, generates a response and sends the response to user interface 307 for display; [0094-0095], the application service inputs a second prompt to the LLM service which instructs the LLM service to provide a chain-of-thought breakdown or decomposition of the formula generated for the first prompt; [0097-0098], In FIG. 7B, upon receiving the formula in response to the first prompt, prompt engine 305 generates a second prompt which tasks LLM 330 with using Excel's LET function to break the formula down into steps; [0099-0101], FIGS. 8A, 8B, and 8C illustrate operational scenario 800 which presents an alternative presentation of the reply to the second prompt from LLM 330 beginning with, in FIG. 8A, receiving the user input requesting a way to calculate the surface area of a cylinder; Based on the reply from LLM 330 to the second prompt, prompt engine 305 configures and progressively displays the parts or steps of the output in user interface 307. As each part is displayed, the user can manually advance to the disclosure of the next part, for example, by clicking an “Accept.”; [0102-0103], In FIG. 9A, the user selects a cell next to the last column of the data table. The application floats a suggestion to allow the application to create a column for the user; [0107-0111], In FIG. 10C, task pane 1010 is updated by the application to show input 1016 and output 1017 based on the LLM's reply to the prompt based on input 1016. Output 1017 includes three suggestions received from the LLM in response to input 1016 according to parameters provided by the application in the prompt) Regarding claim 18, Fabian teaches the claim comprising: A non-transitory computer-readable medium storing executable instructions which, when executed by a processing device, cause the processing device to perform operations comprising (Fabian Figs. 1-13; [0119], FIG. 13 illustrates computing device 1301 that is representative of any system or collection of systems in which the various processes, programs, services, and scenarios disclosed herein may be implemented. Examples of computing device 1301 include, but are not limited to, desktop and laptop computers, tablet computers, mobile computers, and wearable devices; [0122], processing system 1302 may comprise a micro-processor and other circuitry that retrieves and executes software 1305 from storage system 130; [0123], Storage system 1303 may comprise any computer readable storage media readable by processing system 1302 and capable of storing software 1305) : receiving, from a client device interacting with a software application, a query regarding a task to be performed via a user interface of the software application, the software application having a set of interactive elements that are accessible via the user interface; determining an environment representation of the user interface; determining one or more previous actions that have been executed in performance of the task (Fabian Figs 1-13; [0082], FIGS. 5B-5F continue operational scenario 500, illustrating a turn-based chat including inputs received from a user, prompts generated based on the inputs, and responses based on replies to the inputs from LLM 330. In FIG. 5B, the user enters the natural language inquiry about data table 502 into task pane 503. User interface 307 transmits the user input to prompt engine 305 which generates a prompt based on the input; [0099-0101], FIGS. 8A, 8B, and 8C illustrate operational scenario 800 which presents an alternative presentation of the reply to the second prompt from LLM 330 beginning with, in FIG. 8A, receiving the user input requesting a way to calculate the surface area of a cylinder; [0108], FIGS. 10A, 10B, and 10C illustrate operational scenario 1000 of an LLM integration with a spreadsheet environment; [0032], To generate a prompt based on a user's natural language input, the application will configure the prompt according to one or more of: the scope of the problem, the tasks to be completed, illustrative examples, sample data or spreadsheet contextual information, rules, the output format, and a cue to get the LLM to complete the task and not ramble. The scope of the problem refers to the domain in which the problem is to be addressed. For example, the prompt may instruct the LLM to limit its reply to Microsoft Excel® formulas and not Google Sheets formulas. The tasks to be completed include instructions, such as, “generate a formula to do X and self-evaluate the formula.” Illustrative examples can include an example of how to self-evaluate a formula when the formula is relatively complex. An illustrative example provides the LLM with guidance on how to complete a task of the prompt; [0060], Application service 110 receives the input and generates a prompt to be submitted to LLM service 120 based on the input. The prompt includes contextual information, such as a chat history and spreadsheet data or metadata, such as a table name, worksheet name, or spreadsheet name, row and column headers, and at least a subset of the data, e.g., the first five or ten rows of data. Contextual information can also include the recent or latest actions performed by the user on the spreadsheet, such as the user creating a new column or entering a formula. Contextual information can also include errors detected in the spreadsheet; [0066], To configure the prompt, prompt engine 305 identifies a prompt template according to the type of request in the input in an implementation. Prompt templates can include prompt configurations for suggesting a calculated column to be added to workbook data 320, for a general inquiry about workbook data 320, for analyzing data in workbook data 320 to project a result, such as for a hypothetical scenario, and so on; [0067], Prompt engine 305 configures the prompt including parameters to direct LLM 330 to provide a focused response to the input. Prompt parameters include the scope of the prompt, the output format of the reply to produce a reply in a parse-able format, instructions or tasks, examples including sample data or sample data formatting, special tokens which influence the behavior of LLM 330, and so on; [0082], User interface 307 transmits the user input to prompt engine 305 which generates a prompt based on the input. Prompt engine 305 selects a prompt template based on a type or classification of the inquiry and configures a prompt according to the template. In the prompt, prompt engine 305 includes tasks, instructions, or rules applicable to generating a reply to the input, such as tasking LLM 330 to perform a self-evaluation of its reply and a rule to return the reply in a particular output format which is suitable for parsing; [0083], Prompt engine 305 also includes contextual information in the prompt. Prompt engine 305 retrieves spreadsheet context data relating to data table 502 from various ones of application components 303. For example, for a general inquiry about the contents of data table 502, prompt engine 305 configures the prompt to include all the column headers in data table 502 along with the table name and the first five rows of data) ; generating, using a large language model, an estimated lookahead plan describing one or more remaining actions for performing the task by generating the estimated lookahead plan based on the environment representation, the one or more previous actions, and an execution example that includes an example task and an example action sequence for performing the example task, wherein the one or more remaining actions involve use of one or more interactive elements from the set of interactive elements; and generating, from the estimated lookahead plan using and one or more large language models, instructions to perform a next action for performing the task, the instructions indicating an interactive element selected from the set of interactive elements (Fabian Figs 1-13; [0028], the reply from the LLM includes a specific suggestion for modifying the spreadsheet in the specific way; The LLM, upon receiving the configured prompt, generates a reply which suggests adding a column which computes a total sales quantity based on year-to-date (YTD) sales data along with a spreadsheet formula to calculate the quantity. Based on the reply, the application displays the suggestion along with the formula in the user interface of the spreadsheet. In an implementation, the application generates a preview of the column to be added to the data table based on the suggestion, i.e., a column calculating YTD sales data based on the table data and using the formula provided in the reply of the LLM, which the user may then direct the application to implement; [0048], Application service 110 may display the suggestions according to how LLM service 120 self-evaluated the suggestions, e.g., according to relevance to the input or correctness; [0049], Upon receiving the user's selection of the first suggestion in task pane 144, application service 110 implements the suggestion by adding a column (not shown) to the spreadsheet data; [0060], Application service 110 receives the input and generates a prompt to be submitted to LLM service 120 based on the input. The prompt includes contextual information, such as a chat history and spreadsheet data or metadata, such as a table name, worksheet name, or spreadsheet name, row and column headers, and at least a subset of the data, e.g., the first five or ten rows of data. Contextual information can also include the recent or latest actions performed by the user on the spreadsheet, such as the user creating a new column or entering a formula. Contextual information can also include errors detected in the spreadsheet; [0084], Continuing with FIG. 5B, with the prompt configured, prompt engine 305 sends the prompt to LLM 330. LLM 330 generates a reply to the prompt and transmits the reply to prompt engine 305; [0087], Upon receiving LLM 330's reply, prompt engine 305 post-processes the reply to generate a response for display in user interface 307. Post-processing the reply includes extracting information from the reply according to the output formatting rules, such as a description of the suggestion and instructions by which the suggestion can be implemented either by application components 303 or by the user. Continuing to FIG. 5D, the response includes a link by which the user can view more detailed information about the suggestion and a graphical button by which to add the column to data table 502. The user clicks the “Add column” button, causing one or more of application components 303 to modify data table 502 to include the new column and to fill the column according to instructions provided in the reply; [0088-0089], LLM 330 generates two formulas along with the descriptions and explanations and determines, in FIG. 5F, that the formulas are of moderate and high relevance to the input. LLM 330 transmits its reply to prompt engine 305 which, in turn, generates a response and sends the response to user interface 307 for display; [0094], the application service inputs a second prompt to the LLM service which instructs the LLM service to provide a chain-of-thought breakdown or decomposition of the formula generated for the first prompt; [0095], the second prompt generated by the application service directs the LLM service to return the formula generated in response to the first prompt in terms of the Microsoft Excel LET function; [0097]. FIGS. 7A and 7B illustrate operational scenario 700 of prompting an LLM service by an application service involving the use of chain-of-thought decomposition; [0098], In FIG. 7B, upon receiving the formula in response to the first prompt, prompt engine 305 generates a second prompt which tasks LLM 330 with using Excel's LET function to break the formula down into steps; [0099-0101], FIGS. 8A, 8B, and 8C illustrate operational scenario 800 which presents an alternative presentation of the reply to the second prompt from LLM 330 beginning with, in FIG. 8A, receiving the user input requesting a way to calculate the surface area of a cylinder; Based on the reply from LLM 330 to the second prompt, prompt engine 305 configures and progressively displays the parts or steps of the output in user interface 307. As each part is displayed, the user can manually advance to the disclosure of the next part, for example, by clicking an “Accept.”; [0102], FIGS. 9A-9E illustrate operational scenario 900; [0103], In FIG. 9A, the user selects a cell next to the last column of the data table. The application floats a suggestion to allow the application to create a column for the user; [0107], FIG. 9D illustrates task pane 910 subsequent to the user selecting the Years of Service column for inclusion in the data table. In textbox 911, the user enters a natural language input to “Add a column with the employee's last name.” The application configures a prompt based on the input which tasks the LLM to produce suggestions in response to the input; [0108], FIGS. 10A, 10B, and 10C illustrate operational scenario 1000 of an LLM integration with a spreadsheet environment; [0109], The user is also presented with textbox 1013 by which the user can submit natural language inputs, such as requests or queries; [0110], The LLM generates and sends a reply to the prompt to the application which the application processes to produce a response to the input. The response to the input from the application based on the reply from the LLM is shown in output 1015. Output 1015 includes an explanation of the reply and a table demonstrating a spreadsheet calculation, including spreadsheet formulas, demonstrating a calculation responsive to the input; [0111], In FIG. 10C, task pane 1010 is updated by the application to show input 1016 and output 1017 based on the LLM's reply to the prompt based on input 1016. Output 1017 includes three suggestions received from the LLM in response to input 1016 according to parameters provided by the application in the prompt) However, Fabian fails to expressly disclose the instructions indicating an interactive element selected from the set of interactive elements and prompting user interaction with the interactive element via the user interface in performance of the next action. In the same field of endeavor, Mathur teaches: the instructions indicating an interactive element selected from the set of interactive elements and prompting user interaction with the interactive element via the user interface in performance of the next action (Mathur Figs. 1-7; [0015], the response may miss certain steps that are crucial information for the user in performing an action; the response may need to include a sequential number of steps for performing an action in the application; [0021], The application 120 and local applications 164A-164N (collectively referred to as local application 164) may be any application that enables a user such as users 162A-162N (collectively referred to as user 162) to interact with the application to perform an action or achieve a purpose; [0024], the application 120 and/or application 164 utilize the LLM 130 to provide an application copilot that assists users in navigating application features and/or performing actions within the application; [0036], The path extraction engine 240 examines the response to extract a relevant path of actions included in the response. The terms “action path”, “path of actions” or path as used herein refer to a sequence of terms included in a response from the language model or in a help documentation, where each term refers to an action a user takes or refers to a user interface elements a user selects to take an action to perform a task associated with the product help inquiry., the extracted action path is a sequence of terms used in the response that correspond with specific actions or user interface elements in the product. For example, the action path for the example response depicted in FIG. 4B which explains how to create a new calendar meeting, is “New Meeting”, “Invite People”, and “Send.” The actions correspond with specific user interface elements of the meeting application. The path extraction engine 240 may utilize information about the product/application to retrieve key terms in the response that are associated with product actions/UI elements; [0044], FIGS. 4A-4C depict example user interface (UI) screens of an application that utilizes a language model to provide responses to product help inquiries. FIG. 4A depicts an example graphical user interface (GUI) screen 400A of an application that provides an assistant (e.g., a copilot) for utilizing the application. In an example, the assistant if provided for a communications application such as Microsoft Teams. As depicted, the assistant may begin the conversation (e.g., when a user invokes the assistance) by displaying a UI element 410 that asks how it can help the user. The user can utilize a UI element 412 to enter a user query. In the example depicted in screen 400A, the user seeks guidance in creating a calendar meeting. In response, the assistant (e.g., language model) provides summarized instructions on how to create a calendar meeting using the application by displaying the instructions in a UI element 414; [0045], FIG. 4B depicts an alternative example of displaying the response. The response depicted in the UI element 420 of FIG. 4B includes more detailed instructions on how to create a new meeting in the application; [0051], Once the prompt is transmitted to the model, a response to the product help inquiry is received from the model, at 510. The response may include a set of instructions (e.g., an ordered list of actions) to follow to achieve the results indicated in the product help inquiry. This list of actions may be referred to as a path of actions. Method 500 proceeds to extract this path of actions from the response, at 512. This may involve use of an ML model and/or one or more classifiers and utilizing a resource code for the application to identify commands provided by the application or UI elements used by the application that are included in the response) It would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to have incorporated the instructions indicating an interactive element selected from the set of interactive elements and prompting user interaction with the interactive element via the user interface in performance of the next action as suggested in Mathur into Fabian in view of Mathur. Doing so would be desirable because Recently, there has been a significant increase in the use of machine-learning (ML) models such as large language models (LLM) to provide a variety of services and functions. Many LLMs receive an input such as a text segment and provide a prediction based on the input. Because of the wide use of LLMs, it is important that such models provide accurate results. However, even when aggregate accuracy is high, LLMs often fail when providing responses relating to specific domains such as specific products or topics (see Mathur [0001]). Hence, there is a need for improved systems and methods of evaluating responses provided by LLMs (see Mathur [0002]). A user can ask the chatbot how to perform a specific function within the application (e.g., how do I create a calendar event?). This provides an easy-to-use interface by which users can determine how to use an application or any other product. However, in order to ensure that the LLM provides accurate responses to user questions, the LLM would need finetuning with information related to the specific application or product. Finetuning an LLM, however, is often an expensive undertaking. That is because when focusing on a specific domain, it is often difficult to test and engineer prompts in a way that is scalable and has sufficient samples. Moreover, even a finetuned LLM may still exhibit failures in specific domains of data. For example, the LLM may provide harmful, inaccurate or irrelevant responses to specific queries. In an example, the response given might be written well but might have incomplete knowledge of the query. In another example, the response may miss certain steps that are crucial information for the user in performing an action. Such responses result in user frustration, inefficiency and erosion of user trust in the product. Thus, when considering integrations and implementations of LLMs in various tools and applications, it is important to have a reliable metric for evaluating the model's success and the specificity of the model to a specific domain (e.g., specific product). This is particularly true for an LLM that is finetuned to provide responses to product help inquiries, where the response may need to include a sequential number of steps for performing an action in the application (see Mathur [0015]). The system of Mathur can be used with a variety of applications, such as a communications application (e.g., Microsoft® Teams®), presentation application, design application, word processing application, spreadsheet application, social media application, and any other application for which a user may require help (see Mathur [0021]). Additionally, the system of Mathur would improve the system of Fabian by clarifying that the user may manually implement the steps provided in a response (see Fabian Figs. 7-10). Allowing the user to determine how to proceed based on the response would provide improved flexibility in situations where the user may prefer to perform an action themselves rather than rely on the system, which would increase ease of use and user satisfaction. Regarding claim 19, Fabian in view of Mathur teaches all the limitations of claim 18, further comprising: wherein generating, from the estimated lookahead plan and using the one or more large language models, the instructions to perform the next action comprises generating, from a next operation indicated by the estimated lookahead plan and using the one or more large language models, the instructions to perform the next action (Fabian Figs 1-13; [0028], the reply from the LLM includes a specific suggestion for modifying the spreadsheet in the specific way; The LLM, upon receiving the configured prompt, generates a reply which suggests adding a column which computes a total sales quantity based on year-to-date (YTD) sales data along with a spreadsheet formula to calculate the quantity; the user may then direct the application to implement; [0032], To generate a prompt based on a user's natural language input, the application will configure the prompt according to one or more of: the scope of the problem, the tasks to be completed, illustrative examples, sample data or spreadsheet contextual information, rules, the output format, and a cue to get the LLM to complete the task and not ramble. The scope of the problem refers to the domain in which the problem is to be addressed. For example, the prompt may instruct the LLM to limit its reply to Microsoft Excel® formulas and not Google Sheets formulas. The tasks to be completed include instructions, such as, “generate a formula to do X and self-evaluate the formula.”; [0048-0049], Upon receiving the user's selection of the first suggestion in task pane 144, application service 110 implements the suggestion by adding a column (not shown) to the spreadsheet data; [0084-0087], The user clicks the “Add column” button, causing one or more of application components 303 to modify data table 502 to include the new column and to fill the column according to instructions provided in the reply; [0088-0089], LLM 330 generates two formulas along with the descriptions and explanations and determines, in FIG. 5F, that the formulas are of moderate and high relevance to the input. LLM 330 transmits its reply to prompt engine 305 which, in turn, generates a response and sends the response to user interface 307 for display; [0094-0095], the application service inputs a second prompt to the LLM service which instructs the LLM service to provide a chain-of-thought breakdown or decomposition of the formula generated for the first prompt; [0097-0098], In FIG. 7B, upon receiving the formula in response to the first prompt, prompt engine 305 generates a second prompt which tasks LLM 330 with using Excel's LET function to break the formula down into steps; [0099-0101], FIGS. 8A, 8B, and 8C illustrate operational scenario 800 which presents an alternative presentation of the reply to the second prompt from LLM 330 beginning with, in FIG. 8A, receiving the user input requesting a way to calculate the surface area of a cylinder; Based on the reply from LLM 330 to the second prompt, prompt engine 305 configures and progressively displays the parts or steps of the output in user interface 307. As each part is displayed, the user can manually advance to the disclosure of the next part, for example, by clicking an “Accept.”; [0102-0103], In FIG. 9A, the user selects a cell next to the last column of the data table. The application floats a suggestion to allow the application to create a column for the user; [0107-0111], In FIG. 10C, task pane 1010 is updated by the application to show input 1016 and output 1017 based on the LLM's reply to the prompt based on input 1016. Output 1017 includes three suggestions received from the LLM in response to input 1016 according to parameters provided by the application in the prompt) Regarding claim 20, Fabian in view of Mathur teaches all the limitations of claim 19, further comprising: wherein generating, from the next operation indicated by the estimated lookahead plan and using the one or more large language models, the instructions to perform the next action comprises: generating, using an additional large language model, an operation for the next action; and generating the instructions to include the operation as part of the next action based on comparing the operation generated via the additional large language model to the next operation indicated by the estimated lookahead plan (Fabian Figs 1-13; [0027], the user's general inquiry is ambiguous or underspecified, and the application prompts the LLM to interpret the reply in multiple ways and to generate suggestions based on the multiple interpretations; The application may include in its prompt to the LLM an instruction to interpret the inquiry in multiple ways and to generate suggestions based on the interpretations; [0028], the reply from the LLM includes a specific suggestion for modifying the spreadsheet in the specific way; The LLM, upon receiving the configured prompt, generates a reply which suggests adding a column which computes a total sales quantity based on year-to-date (YTD) sales data along with a spreadsheet formula to calculate the quantity; the user may then direct the application to implement; [0032], To generate a prompt based on a user's natural language input, the application will configure the prompt according to one or more of: the scope of the problem, the tasks to be completed, illustrative examples, sample data or spreadsheet contextual information, rules, the output format, and a cue to get the LLM to complete the task and not ramble. The scope of the problem refers to the domain in which the problem is to be addressed. For example, the prompt may instruct the LLM to limit its reply to Microsoft Excel® formulas and not Google Sheets formulas. The tasks to be completed include instructions, such as, “generate a formula to do X and self-evaluate the formula.” Illustrative examples can include an example of how to self-evaluate a formula when the formula is relatively complex; [0036], The application may order the alternative suggestions or eliminate suggestions based on their self-evaluations and present the alternative suggestions in the task pane of the user interface based on the self-evaluations; [0037], if the user request is determined to be ambiguous, the application may direct the LLM to produce a formula and suggest alternative interpretations of the query; [0048-0049], suggested actions may be based on a higher-ranked suggestion received from LLM service 120 in response to the user's original input; [0084-0087], the user submits a second input in task pane 503 requesting suggestions for making “it” better. Prompt engine 305 selects a template to configure a prompt. For an ambiguous or underspecified input, the prompt template may include an instruction to LLM 330 to interpret the input in multiple ways and to generate multiple alternative suggestions based on the interpretations. Here, “it” may refer to, for example, the formatting of data table 502, the organization of data table 502, to the contents of data table 502, or to some other aspect of data table 502. Prompt engine 305 adds context information to the prompt: relevant spreadsheet data, such as the column headers and some rows of data, and the chat history thus far. The prompt also includes an instruction for LLM 330 to self-evaluate its suggestions and to filter out suggestions which are of less than moderate relevance to the subject of the inquiry. The tasks may also include a rule to generate, for any suggested formulas produced by LLM 330, a description and/or explanation of the formula enclosed in the appropriate tags; [0088-0089], LLM 330 generates two formulas along with the descriptions and explanations and determines, in FIG. 5F, that the formulas are of moderate and high relevance to the input. LLM 330 transmits its reply to prompt engine 305 which, in turn, generates a response and sends the response to user interface 307 for display; [0094], the application service inputs a second prompt to the LLM service which instructs the LLM service to provide a chain-of-thought breakdown or decomposition of the formula generated for the first prompt; [0095], In an implementation of a chain-of-thought breakdown, the second prompt generated by the application service directs the LLM service to return the formula generated in response to the first prompt in terms of the Microsoft Excel LET function; [0096], The second formula, in the same programming language as the first formula, is a chain-of-thought decomposition of the first formula into multiple parts such that executing the multiple parts in sequence effects the operation of the first formula; [0097], FIGS. 7A and 7B illustrate operational scenario 700 of prompting an LLM service by an application service involving the use of chain-of-thought decomposition; In FIG. 7B, upon receiving the formula in response to the first prompt, prompt engine 305 generates a second prompt which tasks LLM 330 with using Excel's LET function to break the formula down into steps; [0099-0101], FIGS. 8A, 8B, and 8C illustrate operational scenario 800 which presents an alternative presentation of the reply to the second prompt from LLM 330 beginning with, in FIG. 8A, receiving the user input requesting a way to calculate the surface area of a cylinder; Based on the reply from LLM 330 to the second prompt, prompt engine 305 configures and progressively displays the parts or steps of the output in user interface 307. As each part is displayed, the user can manually advance to the disclosure of the next part, for example, by clicking an “Accept.”; [0102-0103], In FIG. 9A, the user selects a cell next to the last column of the data table. The application floats a suggestion to allow the application to create a column for the user; [0107-0111], In FIG. 10C, task pane 1010 is updated by the application to show input 1016 and output 1017 based on the LLM's reply to the prompt based on input 1016. Output 1017 includes three suggestions received from the LLM in response to input 1016 according to parameters provided by the application in the prompt) 07-21-aia AIA Claim s 5, 6, 12-14 are rejected under 35 U.S.C. 103 as being unpatentable over Fabian in view of Mathur in further view of Ghaeini et al. (US 20250111202 A1, published 04/03/2025), hereinafter Ghaeini . Regarding claim 5, Fabian in view of Mathur teaches all the limitations of claim 1, further comprising: generating, using an encoding model, an encoding of the query; generating, using the encoding model, a plurality of additional encodings for a plurality of execution examples that are from a dataset of execution examples; selecting the at least one execution example from the plurality of execution examples to include in the lookahead prompt (Fabian Figs 1-13; [0028], the reply from the LLM includes a specific suggestion for modifying the spreadsheet in the specific way; The LLM, upon receiving the configured prompt, generates a reply which suggests adding a column which computes a total sales quantity based on year-to-date (YTD) sales data along with a spreadsheet formula to calculate the quantity; the user may then direct the application to implement; [0032], To generate a prompt based on a user's natural language input, the application will configure the prompt according to one or more of: the scope of the problem, the tasks to be completed, illustrative examples, sample data or spreadsheet contextual information, rules, the output format, and a cue to get the LLM to complete the task and not ramble; [0067], Prompt engine 305 configures the prompt including parameters to direct LLM 330 to provide a focused response to the input. Prompt parameters include the scope of the prompt, the output format of the reply to produce a reply in a parse-able format, instructions or tasks, examples including sample data or sample data formatting, special tokens which influence the behavior of LLM 330, and so on; [0048-0049], Upon receiving the user's selection of the first suggestion in task pane 144, application service 110 implements the suggestion by adding a column (not shown) to the spreadsheet data; [0084-0087], The user clicks the “Add column” button, causing one or more of application components 303 to modify data table 502 to include the new column and to fill the column according to instructions provided in the reply; [0088-0089], LLM 330 generates two formulas along with the descriptions and explanations and determines, in FIG. 5F, that the formulas are of moderate and high relevance to the input. LLM 330 transmits its reply to prompt engine 305 which, in turn, generates a response and sends the response to user interface 307 for display; see also [0094-0103], [0107-0111]) However, Fabian in view of Mathur fails to expressly disclose generating, using an encoding model, an encoding of the query; generating, using the encoding model, a plurality of additional encodings for a plurality of execution examples that are from a dataset of execution examples; determining pairwise cosine similarities between the encoding of the query and the plurality of additional encodings; and selecting the at least one execution example from the plurality of execution examples to include in the lookahead prompt based on the pairwise cosine similarities. In the same field of endeavor, Ghaeini teaches: generating, using an encoding model, an encoding of the query; generating, using the encoding model, a plurality of additional encodings for a plurality of execution examples that are from a dataset of execution examples; determining pairwise cosine similarities between the encoding of the query and the plurality of additional encodings; and selecting the at least one execution example from the plurality of execution examples to include in the lookahead prompt based on the pairwise cosine similarities (Ghaeini Figs. 1-5; [0013], The technology disclosed herein, among other things, provides solutions to the above problem by providing systems and methods that dynamically generate prompts based on the input content. The dynamic generation of the prompts results in prompts that are more computationally efficient while preserving the clarity and quality of the prompts. With the dynamic prompt generation, the input content is preprocessed to determine which categories and/or examples are most closely related to the input content. Based on that similarity determination, only the examples and/or categories that are determined to be most closely related (e.g., exceeding a similarity metric) are incorporated into the prompt; [0018], the system 100 further includes a dynamic prompt generator 110 that dynamically generates prompts, as discussed herein. For example, the dynamic prompt generator 110 receives input content from the content application 108 that is to be evaluated (e.g., classified). Based on the input content, the dynamic prompt generator 110 generates a dynamic prompt; [0020], The trait repository 118 includes example pre-tagged data, which may also be referred to herein as trait data. For instance, the trait data may include content that has already been tagged with a known classification tag or label; [0037], embedding requestor 252 generates a request for an embedding to be created for the input content 280. The embedding requestor 252 transmits the embedding request to the embedding generator 116, where the embedding generator 116 generates an embedding for the input content 280; [0038], the embedding requestor 252 also generates a request for embeddings of the trait data stored in the trait repository 118; [0039], The embedding comparer 254 then compares the input-content embedding with the trait-data embedding to identify trait data that is similar to the input content 280. In some examples, comparison of the input-content embedding and the trait-data embedding is performed as a cosine similarity analysis performed over the vector space of the embeddings. For instance, the top N number of trait data may be identified based on the comparison. In such examples, the output of the embedding comparer 254 is a ranked list of trait data, where the ranking of the trait data is based on the similarity of the trait data to the input content 280. In other examples, the trait data that exceeds a similarity threshold when compared to the input data is identified by the embedding comparer 254 as similar to the input content 280; [0040], Based on the trait data that is found to be similar to the input content 280 (e.g., the trait data having the highest similarity with the input content 280), the prompt builder 256 builds a prompt; [0064], The cosine similarity analysis of the embeddings (which are multidimensional vectors) provides a measure of how close the embeddings are in the multidimensional space. In such examples, the comparison of each trait-data embedding and the input-content embedding results in a similarity score, which may be a cosine similarity score (e.g., −1 to 1 with a score of 1 indicating an identical vector); [0067], a different task (e.g., a different classification task or another task altogether), a second or different prompt template is accessed that relates to the particular evaluation task) It would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to have incorporated generating, using an encoding model, an encoding of the query; generating, using the encoding model, a plurality of additional encodings for a plurality of execution examples that are from a dataset of execution examples; determining pairwise cosine similarities between the encoding of the query and the plurality of additional encodings; and selecting the at least one execution example from the plurality of execution examples to include in the lookahead prompt based on the pairwise cosine similarities as suggested in Ghaeini into Fabian in view of Mathur. Doing so would be desirable because the use of LLMs provides solutions for a wide variety of applications. However, LLMs are also resource-intensive solutions. LLMs are often configured to process a prompt that may include natural language instructions and/or requests for the LLM to process (see Ghaeini [0010]). The size and configuration of the prompt also affects the performance of the LLM. Shorter prompts provide for faster processing of the prompt and/or a smaller memory footprint. Nevertheless, reducing the length of the prompt may require the omission of data that may have otherwise improved the clarity and/or quality of the prompt (see Ghaeini [0011]). Prompts may include explicit examples that help guide the LLM to provide a more accurate response. There are a vast number of examples of misinformation that can be compiled and tagged. Thus, incorporating all the possible examples into the prompt increases the computing resources required to process the prompt by the LLM. The increased computing resources also often leads to increased latency in generating the response (see Ghaeini [0012]). The technology disclosed herein, among other things, provides solutions to the above problem by providing systems and methods that dynamically generate prompts based on the input content. The dynamic generation of the prompts results in prompts that are more computationally efficient while preserving the clarity and quality of the prompts. With the dynamic prompt generation, the input content is preprocessed to determine which categories and/or examples are most closely related to the input content. Based on that similarity determination, only the examples and/or categories that are determined to be most closely related (e.g., exceeding a similarity metric) are incorporated into the prompt. This allows for the examples and/or categories that are least likely to be useful for evaluation of the input content to be omitted from the prompt. Accordingly, the dynamically generated prompt allows for improved computational performance by the LLM (when the LLM processes the prompt) while still retaining the data that is most likely to lead to an accurate evaluation of the input content (see Ghaeini [0013]). Additionally, the system of Ghaeini can be used for a variety of tasks (see Ghaeini [0067]). Regarding claim 6, Fabian in view of Mathur in further view of Ghaeini teaches all the limitations of claim 5, further comprising: generating the plurality of additional encodings for the plurality of execution examples comprises generating the plurality of additional encodings for a set of execution examples associated with a plurality of software applications; and selecting the at least one execution example to include in the lookahead prompt comprises selecting the at least one execution example based on determining that the at least one execution example is associated with the software application (Fabian Figs 1-13; [0028], the reply from the LLM includes a specific suggestion for modifying the spreadsheet in the specific way; The LLM, upon receiving the configured prompt, generates a reply which suggests adding a column which computes a total sales quantity based on year-to-date (YTD) sales data along with a spreadsheet formula to calculate the quantity; the user may then direct the application to implement; [0032], To generate a prompt based on a user's natural language input, the application will configure the prompt according to one or more of: the scope of the problem, the tasks to be completed, illustrative examples, sample data or spreadsheet contextual information, rules, the output format, and a cue to get the LLM to complete the task and not ramble. The scope of the problem refers to the domain in which the problem is to be addressed. For example, the prompt may instruct the LLM to limit its reply to Microsoft Excel® formulas and not Google Sheets formulas; [0043], Operational environment 100 includes application service 110, LLM service 120, and computing devices 130. Application service 110 hosts a productivity application such as a spreadsheet application (e.g., Microsoft Excel) to endpoints such as computing devices 130 which execute applications that provide a local user experience and that interface with application service 110. The applications running locally with respect to computing devices 130 may be natively installed and executed applications, browser-based applications, mobile applications, streamed applications; [0051], Process 200 may be implemented in program instructions in the context of any of the software applications; [0067], Prompt engine 305 configures the prompt including parameters to direct LLM 330 to provide a focused response to the input. Prompt parameters include the scope of the prompt, the output format of the reply to produce a reply in a parse-able format, instructions or tasks, examples including sample data or sample data formatting, special tokens which influence the behavior of LLM 330, and so on; [0048-0049], Upon receiving the user's selection of the first suggestion in task pane 144, application service 110 implements the suggestion by adding a column (not shown) to the spreadsheet data; [0084-0087], The user clicks the “Add column” button, causing one or more of application components 303 to modify data table 502 to include the new column and to fill the column according to instructions provided in the reply; [0088-0089], LLM 330 generates two formulas along with the descriptions and explanations and determines, in FIG. 5F, that the formulas are of moderate and high relevance to the input. LLM 330 transmits its reply to prompt engine 305 which, in turn, generates a response and sends the response to user interface 307 for display; [0091], Process 600 may be implemented in program instructions in the context of any of the software applications; see also [0094-0103], [0107-0111]) Ghaeini further teaches: generating the plurality of additional encodings for the plurality of execution examples comprises generating the plurality of additional encodings for a set of execution examples associated with a plurality of software applications; and selecting the at least one execution example to include in the lookahead prompt based on the pairwise cosine similarities comprises selecting the at least one execution example based on the pairwise cosine similarities and based on determining that the at least one execution example is associated with the software application (Ghaeini Figs. 1-5; [0013], The technology disclosed herein, among other things, provides solutions to the above problem by providing systems and methods that dynamically generate prompts based on the input content. The dynamic generation of the prompts results in prompts that are more computationally efficient while preserving the clarity and quality of the prompts. With the dynamic prompt generation, the input content is preprocessed to determine which categories and/or examples are most closely related to the input content. Based on that similarity determination, only the examples and/or categories that are determined to be most closely related (e.g., exceeding a similarity metric) are incorporated into the prompt; [0014], components of the system 100 are illustrative of software applications; [0017], content application(s) 108 may be local applications or web-based applications; [0018], the system 100 further includes a dynamic prompt generator 110 that dynamically generates prompts, as discussed herein. For example, the dynamic prompt generator 110 receives input content from the content application 108 that is to be evaluated (e.g., classified). Based on the input content, the dynamic prompt generator 110 generates a dynamic prompt; [0020], The trait repository 118 includes example pre-tagged data, which may also be referred to herein as trait data. For instance, the trait data may include content that has already been tagged with a known classification tag or label; [0037], embedding requestor 252 generates a request for an embedding to be created for the input content 280. The embedding requestor 252 transmits the embedding request to the embedding generator 116, where the embedding generator 116 generates an embedding for the input content 280; [0038], the embedding requestor 252 also generates a request for embeddings of the trait data stored in the trait repository 118; [0039], The embedding comparer 254 then compares the input-content embedding with the trait-data embedding to identify trait data that is similar to the input content 280. In some examples, comparison of the input-content embedding and the trait-data embedding is performed as a cosine similarity analysis performed over the vector space of the embeddings. For instance, the top N number of trait data may be identified based on the comparison. In such examples, the output of the embedding comparer 254 is a ranked list of trait data, where the ranking of the trait data is based on the similarity of the trait data to the input content 280. In other examples, the trait data that exceeds a similarity threshold when compared to the input data is identified by the embedding comparer 254 as similar to the input content 280; [0040], Based on the trait data that is found to be similar to the input content 280 (e.g., the trait data having the highest similarity with the input content 280), the prompt builder 256 builds a prompt; [0064], The cosine similarity analysis of the embeddings (which are multidimensional vectors) provides a measure of how close the embeddings are in the multidimensional space. In such examples, the comparison of each trait-data embedding and the input-content embedding results in a similarity score, which may be a cosine similarity score (e.g., −1 to 1 with a score of 1 indicating an identical vector); [0067], a different task (e.g., a different classification task or another task altogether), a second or different prompt template is accessed that relates to the particular evaluation task) It would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to have incorporated generating the plurality of additional encodings for the plurality of execution examples comprises generating the plurality of additional encodings for a set of execution examples associated with a plurality of software applications; and selecting the at least one execution example to include in the lookahead prompt based on the pairwise cosine similarities comprises selecting the at least one execution example based on the pairwise cosine similarities and based on determining that the at least one execution example is associated with the software application as suggested in Ghaeini into Fabian in view of Mathur. Doing so would be desirable because the use of LLMs provides solutions for a wide variety of applications. However, LLMs are also resource-intensive solutions. LLMs are often configured to process a prompt that may include natural language instructions and/or requests for the LLM to process (see Ghaeini [0010]). The size and configuration of the prompt also affects the performance of the LLM. Shorter prompts provide for faster processing of the prompt and/or a smaller memory footprint. Nevertheless, reducing the length of the prompt may require the omission of data that may have otherwise improved the clarity and/or quality of the prompt (see Ghaeini [0011]). Prompts may include explicit examples that help guide the LLM to provide a more accurate response. There are a vast number of examples of misinformation that can be compiled and tagged. Thus, incorporating all the possible examples into the prompt increases the computing resources required to process the prompt by the LLM. The increased computing resources also often leads to increased latency in generating the response (see Ghaeini [0012]). The technology disclosed herein, among other things, provides solutions to the above problem by providing systems and methods that dynamically generate prompts based on the input content. The dynamic generation of the prompts results in prompts that are more computationally efficient while preserving the clarity and quality of the prompts. With the dynamic prompt generation, the input content is preprocessed to determine which categories and/or examples are most closely related to the input content. Based on that similarity determination, only the examples and/or categories that are determined to be most closely related (e.g., exceeding a similarity metric) are incorporated into the prompt. This allows for the examples and/or categories that are least likely to be useful for evaluation of the input content to be omitted from the prompt. Accordingly, the dynamically generated prompt allows for improved computational performance by the LLM (when the LLM processes the prompt) while still retaining the data that is most likely to lead to an accurate evaluation of the input content (see Ghaeini [0013]). Additionally, the system of Ghaeini can be used for a variety of tasks (see Ghaeini [0067]). Regarding claim 12, Fabian in view of Mathur teaches all the limitations of claim 10, further comprising: wherein the one or more processors are configured to cause the system to generate the estimated lookahead plan from the set of candidate interactive elements using the first large language model by: determining a set of execution examples corresponding to the task, each execution example from the set of execution examples including an example task and an example action sequence for performing the example task; generating associations between an encoding of the query and a plurality of additional encodings of the set of execution examples; and generating, using the first large language model, the estimated lookahead plan from the set of candidate interactive elements and one or more execution examples, wherein the one or more execution examples are selected based on the associations (Fabian Figs 1-13; [0028], the reply from the LLM includes a specific suggestion for modifying the spreadsheet in the specific way; The LLM, upon receiving the configured prompt, generates a reply which suggests adding a column which computes a total sales quantity based on year-to-date (YTD) sales data along with a spreadsheet formula to calculate the quantity; the user may then direct the application to implement; [0032], To generate a prompt based on a user's natural language input, the application will configure the prompt according to one or more of: the scope of the problem, the tasks to be completed, illustrative examples, sample data or spreadsheet contextual information, rules, the output format, and a cue to get the LLM to complete the task and not ramble; [0067], Prompt engine 305 configures the prompt including parameters to direct LLM 330 to provide a focused response to the input. Prompt parameters include the scope of the prompt, the output format of the reply to produce a reply in a parse-able format, instructions or tasks, examples including sample data or sample data formatting, special tokens which influence the behavior of LLM 330, and so on; [0048-0049], Upon receiving the user's selection of the first suggestion in task pane 144, application service 110 implements the suggestion by adding a column (not shown) to the spreadsheet data; [0084-0087], The user clicks the “Add column” button, causing one or more of application components 303 to modify data table 502 to include the new column and to fill the column according to instructions provided in the reply; [0088-0089], LLM 330 generates two formulas along with the descriptions and explanations and determines, in FIG. 5F, that the formulas are of moderate and high relevance to the input. LLM 330 transmits its reply to prompt engine 305 which, in turn, generates a response and sends the response to user interface 307 for display; see also [0094-0103], [0107-0111]) However, Fabian in view of Mathur fails to expressly disclose generating pairwise cosine similarities between an encoding of the query and a plurality of additional encodings of the set of execution examples; and generating, using the first large language model, the estimated lookahead plan from the set of candidate interactive elements and one or more execution examples selected from the set of execution examples, wherein the one or more execution examples are selected based on the pairwise cosine similarities. In the same field of endeavor, Ghaeini teaches: generating pairwise cosine similarities between an encoding of the query and a plurality of additional encodings of the set of execution examples; and generating, using the first large language model, the estimated lookahead plan from the set of candidate interactive elements and one or more execution examples selected from the set of execution examples, wherein the one or more execution examples are selected based on the pairwise cosine similarities (Ghaeini Figs. 1-5; [0013], The technology disclosed herein, among other things, provides solutions to the above problem by providing systems and methods that dynamically generate prompts based on the input content. The dynamic generation of the prompts results in prompts that are more computationally efficient while preserving the clarity and quality of the prompts. With the dynamic prompt generation, the input content is preprocessed to determine which categories and/or examples are most closely related to the input content. Based on that similarity determination, only the examples and/or categories that are determined to be most closely related (e.g., exceeding a similarity metric) are incorporated into the prompt; [0018], the system 100 further includes a dynamic prompt generator 110 that dynamically generates prompts, as discussed herein. For example, the dynamic prompt generator 110 receives input content from the content application 108 that is to be evaluated (e.g., classified). Based on the input content, the dynamic prompt generator 110 generates a dynamic prompt; [0020], The trait repository 118 includes example pre-tagged data, which may also be referred to herein as trait data. For instance, the trait data may include content that has already been tagged with a known classification tag or label; [0037], embedding requestor 252 generates a request for an embedding to be created for the input content 280. The embedding requestor 252 transmits the embedding request to the embedding generator 116, where the embedding generator 116 generates an embedding for the input content 280; [0038], the embedding requestor 252 also generates a request for embeddings of the trait data stored in the trait repository 118; [0039], The embedding comparer 254 then compares the input-content embedding with the trait-data embedding to identify trait data that is similar to the input content 280. In some examples, comparison of the input-content embedding and the trait-data embedding is performed as a cosine similarity analysis performed over the vector space of the embeddings. For instance, the top N number of trait data may be identified based on the comparison. In such examples, the output of the embedding comparer 254 is a ranked list of trait data, where the ranking of the trait data is based on the similarity of the trait data to the input content 280. In other examples, the trait data that exceeds a similarity threshold when compared to the input data is identified by the embedding comparer 254 as similar to the input content 280; [0040], Based on the trait data that is found to be similar to the input content 280 (e.g., the trait data having the highest similarity with the input content 280), the prompt builder 256 builds a prompt; [0064], The cosine similarity analysis of the embeddings (which are multidimensional vectors) provides a measure of how close the embeddings are in the multidimensional space. In such examples, the comparison of each trait-data embedding and the input-content embedding results in a similarity score, which may be a cosine similarity score (e.g., −1 to 1 with a score of 1 indicating an identical vector); [0067], a different task (e.g., a different classification task or another task altogether), a second or different prompt template is accessed that relates to the particular evaluation task) It would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to have incorporated generating pairwise cosine similarities between an encoding of the query and a plurality of additional encodings of the set of execution examples; and generating, using the first large language model, the estimated lookahead plan from the set of candidate interactive elements and one or more execution examples selected from the set of execution examples, wherein the one or more execution examples are selected based on the pairwise cosine similarities as suggested in Ghaeini into Fabian in view of Mathur. Doing so would be desirable because the use of LLMs provides solutions for a wide variety of applications. However, LLMs are also resource-intensive solutions. LLMs are often configured to process a prompt that may include natural language instructions and/or requests for the LLM to process (see Ghaeini [0010]). The size and configuration of the prompt also affects the performance of the LLM. Shorter prompts provide for faster processing of the prompt and/or a smaller memory footprint. Nevertheless, reducing the length of the prompt may require the omission of data that may have otherwise improved the clarity and/or quality of the prompt (see Ghaeini [0011]). Prompts may include explicit examples that help guide the LLM to provide a more accurate response. There are a vast number of examples of misinformation that can be compiled and tagged. Thus, incorporating all the possible examples into the prompt increases the computing resources required to process the prompt by the LLM. The increased computing resources also often leads to increased latency in generating the response (see Ghaeini [0012]). The technology disclosed herein, among other things, provides solutions to the above problem by providing systems and methods that dynamically generate prompts based on the input content. The dynamic generation of the prompts results in prompts that are more computationally efficient while preserving the clarity and quality of the prompts. With the dynamic prompt generation, the input content is preprocessed to determine which categories and/or examples are most closely related to the input content. Based on that similarity determination, only the examples and/or categories that are determined to be most closely related (e.g., exceeding a similarity metric) are incorporated into the prompt. This allows for the examples and/or categories that are least likely to be useful for evaluation of the input content to be omitted from the prompt. Accordingly, the dynamically generated prompt allows for improved computational performance by the LLM (when the LLM processes the prompt) while still retaining the data that is most likely to lead to an accurate evaluation of the input content (see Ghaeini [0013]). Additionally, the system of Ghaeini can be used for a variety of tasks (see Ghaeini [0067]). Regarding claim 13, Fabian in view of Mathur in further view of Ghaeini teaches all the limitations of claim 12, further comprising: wherein the one or more processors are further configured to cause the system to select the one or more execution examples from the set of execution examples based on the associations by: determining that one or more associations determined for the one or more execution examples; or determining that the one or more associations determined for the one or more execution examples indicate a association to the query than remaining execution examples from the set of execution examples (Fabian Figs 1-13; [0028], the reply from the LLM includes a specific suggestion for modifying the spreadsheet in the specific way; The LLM, upon receiving the configured prompt, generates a reply which suggests adding a column which computes a total sales quantity based on year-to-date (YTD) sales data along with a spreadsheet formula to calculate the quantity; the user may then direct the application to implement; [0032], the prompt may instruct the LLM to limit its reply to Microsoft Excel® formulas and not Google Sheets formulas; [0043], Operational environment 100 includes application service 110, LLM service 120, and computing devices 130. Application service 110 hosts a productivity application such as a spreadsheet application (e.g., Microsoft Excel) to endpoints such as computing devices 130 which execute applications that provide a local user experience and that interface with application service 110. The applications running locally with respect to computing devices 130 may be natively installed and executed applications, browser-based applications, mobile applications, streamed applications; [0051], Process 200 may be implemented in program instructions in the context of any of the software applications; [0067], Prompt engine 305 configures the prompt including parameters to direct LLM 330 to provide a focused response to the input. Prompt parameters include the scope of the prompt, the output format of the reply to produce a reply in a parse-able format, instructions or tasks, examples including sample data or sample data formatting, special tokens which influence the behavior of LLM 330, and so on; [0048-0049], Upon receiving the user's selection of the first suggestion in task pane 144, application service 110 implements the suggestion by adding a column (not shown) to the spreadsheet data; [0084-0087], The user clicks the “Add column” button, causing one or more of application components 303 to modify data table 502 to include the new column and to fill the column according to instructions provided in the reply; [0088-0089], LLM 330 generates two formulas along with the descriptions and explanations and determines, in FIG. 5F, that the formulas are of moderate and high relevance to the input. LLM 330 transmits its reply to prompt engine 305 which, in turn, generates a response and sends the response to user interface 307 for display; [0091], Process 600 may be implemented in program instructions in the context of any of the software applications; see also [0094-0103], [0107-0111]) Ghaeini further teaches: wherein the one or more processors are further configured to cause the system to select the one or more execution examples from the set of execution examples based on the pairwise cosine similarities by: determining that one or more pairwise cosine similarities determined for the one or more execution examples satisfy a pairwise cosine similarity threshold; or determining that the one or more pairwise cosine similarities determined for the one or more execution examples indicate a higher similarity to the query than remaining execution examples from the set of execution examples (Ghaeini Figs. 1-5; [0013], The technology disclosed herein, among other things, provides solutions to the above problem by providing systems and methods that dynamically generate prompts based on the input content. The dynamic generation of the prompts results in prompts that are more computationally efficient while preserving the clarity and quality of the prompts. With the dynamic prompt generation, the input content is preprocessed to determine which categories and/or examples are most closely related to the input content. Based on that similarity determination, only the examples and/or categories that are determined to be most closely related (e.g., exceeding a similarity metric) are incorporated into the prompt; [0014], components of the system 100 are illustrative of software applications; [0017], content application(s) 108 may be local applications or web-based applications; [0018], the system 100 further includes a dynamic prompt generator 110 that dynamically generates prompts, as discussed herein. For example, the dynamic prompt generator 110 receives input content from the content application 108 that is to be evaluated (e.g., classified). Based on the input content, the dynamic prompt generator 110 generates a dynamic prompt; [0020], The trait repository 118 includes example pre-tagged data, which may also be referred to herein as trait data. For instance, the trait data may include content that has already been tagged with a known classification tag or label; [0037], embedding requestor 252 generates a request for an embedding to be created for the input content 280. The embedding requestor 252 transmits the embedding request to the embedding generator 116, where the embedding generator 116 generates an embedding for the input content 280; [0038], the embedding requestor 252 also generates a request for embeddings of the trait data stored in the trait repository 118; [0039], The embedding comparer 254 then compares the input-content embedding with the trait-data embedding to identify trait data that is similar to the input content 280. In some examples, comparison of the input-content embedding and the trait-data embedding is performed as a cosine similarity analysis performed over the vector space of the embeddings. For instance, the top N number of trait data may be identified based on the comparison. In such examples, the output of the embedding comparer 254 is a ranked list of trait data, where the ranking of the trait data is based on the similarity of the trait data to the input content 280. In other examples, the trait data that exceeds a similarity threshold when compared to the input data is identified by the embedding comparer 254 as similar to the input content 280; [0040], Based on the trait data that is found to be similar to the input content 280 (e.g., the trait data having the highest similarity with the input content 280), the prompt builder 256 builds a prompt; [0064], The cosine similarity analysis of the embeddings (which are multidimensional vectors) provides a measure of how close the embeddings are in the multidimensional space. In such examples, the comparison of each trait-data embedding and the input-content embedding results in a similarity score, which may be a cosine similarity score (e.g., −1 to 1 with a score of 1 indicating an identical vector); [0067], a different task (e.g., a different classification task or another task altogether), a second or different prompt template is accessed that relates to the particular evaluation task) It would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to have incorporated wherein the one or more processors are further configured to cause the system to select the one or more execution examples from the set of execution examples based on the pairwise cosine similarities by: determining that one or more pairwise cosine similarities determined for the one or more execution examples satisfy a pairwise cosine similarity threshold; or determining that the one or more pairwise cosine similarities determined for the one or more execution examples indicate a higher similarity to the query than remaining execution examples from the set of execution examples as suggested in Ghaeini into Fabian in view of Mathur. Doing so would be desirable because the use of LLMs provides solutions for a wide variety of applications. However, LLMs are also resource-intensive solutions. LLMs are often configured to process a prompt that may include natural language instructions and/or requests for the LLM to process (see Ghaeini [0010]). The size and configuration of the prompt also affects the performance of the LLM. Shorter prompts provide for faster processing of the prompt and/or a smaller memory footprint. Nevertheless, reducing the length of the prompt may require the omission of data that may have otherwise improved the clarity and/or quality of the prompt (see Ghaeini [0011]). Prompts may include explicit examples that help guide the LLM to provide a more accurate response. There are a vast number of examples of misinformation that can be compiled and tagged. Thus, incorporating all the possible examples into the prompt increases the computing resources required to process the prompt by the LLM. The increased computing resources also often leads to increased latency in generating the response (see Ghaeini [0012]). The technology disclosed herein, among other things, provides solutions to the above problem by providing systems and methods that dynamically generate prompts based on the input content. The dynamic generation of the prompts results in prompts that are more computationally efficient while preserving the clarity and quality of the prompts. With the dynamic prompt generation, the input content is preprocessed to determine which categories and/or examples are most closely related to the input content. Based on that similarity determination, only the examples and/or categories that are determined to be most closely related (e.g., exceeding a similarity metric) are incorporated into the prompt. This allows for the examples and/or categories that are least likely to be useful for evaluation of the input content to be omitted from the prompt. Accordingly, the dynamically generated prompt allows for improved computational performance by the LLM (when the LLM processes the prompt) while still retaining the data that is most likely to lead to an accurate evaluation of the input content (see Ghaeini [0013]). Additionally, the system of Ghaeini can be used for a variety of tasks (see Ghaeini [0067]). Regarding claim 14, Fabian in view of Mathur in further view of Ghaeini teaches all the limitations of claim 13, further comprising: wherein the one or more processors are further configured to cause the system to select the one or more execution examples from the set of execution examples based the associations by on determining that the one or more execution examples are associated with the software application (Fabian Figs 1-13; [0028], the reply from the LLM includes a specific suggestion for modifying the spreadsheet in the specific way; The LLM, upon receiving the configured prompt, generates a reply which suggests adding a column which computes a total sales quantity based on year-to-date (YTD) sales data along with a spreadsheet formula to calculate the quantity; the user may then direct the application to implement; [0032], To generate a prompt based on a user's natural language input, the application will configure the prompt according to one or more of: the scope of the problem, the tasks to be completed, illustrative examples, sample data or spreadsheet contextual information, rules, the output format, and a cue to get the LLM to complete the task and not ramble. The scope of the problem refers to the domain in which the problem is to be addressed. For example, the prompt may instruct the LLM to limit its reply to Microsoft Excel® formulas and not Google Sheets formulas; [0043], Operational environment 100 includes application service 110, LLM service 120, and computing devices 130. Application service 110 hosts a productivity application such as a spreadsheet application (e.g., Microsoft Excel) to endpoints such as computing devices 130 which execute applications that provide a local user experience and that interface with application service 110. The applications running locally with respect to computing devices 130 may be natively installed and executed applications, browser-based applications, mobile applications, streamed applications; [0051], Process 200 may be implemented in program instructions in the context of any of the software applications; [0067], Prompt engine 305 configures the prompt including parameters to direct LLM 330 to provide a focused response to the input. Prompt parameters include the scope of the prompt, the output format of the reply to produce a reply in a parse-able format, instructions or tasks, examples including sample data or sample data formatting, special tokens which influence the behavior of LLM 330, and so on; [0048-0049], Upon receiving the user's selection of the first suggestion in task pane 144, application service 110 implements the suggestion by adding a column (not shown) to the spreadsheet data; [0084-0087], The user clicks the “Add column” button, causing one or more of application components 303 to modify data table 502 to include the new column and to fill the column according to instructions provided in the reply; [0088-0089], LLM 330 generates two formulas along with the descriptions and explanations and determines, in FIG. 5F, that the formulas are of moderate and high relevance to the input. LLM 330 transmits its reply to prompt engine 305 which, in turn, generates a response and sends the response to user interface 307 for display; [0091], Process 600 may be implemented in program instructions in the context of any of the software applications; see also [0094-0103], [0107-0111]) Ghaeini further teaches: wherein the one or more processors are further configured to cause the system to select the one or more execution examples from the set of execution examples based on the pairwise cosine similarities by determining that the one or more execution examples are associated with the software application (Ghaeini Figs. 1-5; [0013], The technology disclosed herein, among other things, provides solutions to the above problem by providing systems and methods that dynamically generate prompts based on the input content. The dynamic generation of the prompts results in prompts that are more computationally efficient while preserving the clarity and quality of the prompts. With the dynamic prompt generation, the input content is preprocessed to determine which categories and/or examples are most closely related to the input content. Based on that similarity determination, only the examples and/or categories that are determined to be most closely related (e.g., exceeding a similarity metric) are incorporated into the prompt; [0014], components of the system 100 are illustrative of software applications; [0017], content application(s) 108 may be local applications or web-based applications; [0018], the system 100 further includes a dynamic prompt generator 110 that dynamically generates prompts, as discussed herein. For example, the dynamic prompt generator 110 receives input content from the content application 108 that is to be evaluated (e.g., classified). Based on the input content, the dynamic prompt generator 110 generates a dynamic prompt; [0020], The trait repository 118 includes example pre-tagged data, which may also be referred to herein as trait data. For instance, the trait data may include content that has already been tagged with a known classification tag or label; [0037], embedding requestor 252 generates a request for an embedding to be created for the input content 280. The embedding requestor 252 transmits the embedding request to the embedding generator 116, where the embedding generator 116 generates an embedding for the input content 280; [0038], the embedding requestor 252 also generates a request for embeddings of the trait data stored in the trait repository 118; [0039], The embedding comparer 254 then compares the input-content embedding with the trait-data embedding to identify trait data that is similar to the input content 280. In some examples, comparison of the input-content embedding and the trait-data embedding is performed as a cosine similarity analysis performed over the vector space of the embeddings. For instance, the top N number of trait data may be identified based on the comparison. In such examples, the output of the embedding comparer 254 is a ranked list of trait data, where the ranking of the trait data is based on the similarity of the trait data to the input content 280. In other examples, the trait data that exceeds a similarity threshold when compared to the input data is identified by the embedding comparer 254 as similar to the input content 280; [0040], Based on the trait data that is found to be similar to the input content 280 (e.g., the trait data having the highest similarity with the input content 280), the prompt builder 256 builds a prompt; [0064], The cosine similarity analysis of the embeddings (which are multidimensional vectors) provides a measure of how close the embeddings are in the multidimensional space. In such examples, the comparison of each trait-data embedding and the input-content embedding results in a similarity score, which may be a cosine similarity score (e.g., −1 to 1 with a score of 1 indicating an identical vector); [0067], a different task (e.g., a different classification task or another task altogether), a second or different prompt template is accessed that relates to the particular evaluation task) It would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to have incorporated wherein the one or more processors are further configured to cause the system to select the one or more execution examples from the set of execution examples based on the pairwise cosine similarities by determining that the one or more execution examples are associated with the software application as suggested in Ghaeini into Fabian in view of Mathur. Doing so would be desirable because the use of LLMs provides solutions for a wide variety of applications. However, LLMs are also resource-intensive solutions. LLMs are often configured to process a prompt that may include natural language instructions and/or requests for the LLM to process (see Ghaeini [0010]). The size and configuration of the prompt also affects the performance of the LLM. Shorter prompts provide for faster processing of the prompt and/or a smaller memory footprint. Nevertheless, reducing the length of the prompt may require the omission of data that may have otherwise improved the clarity and/or quality of the prompt (see Ghaeini [0011]). Prompts may include explicit examples that help guide the LLM to provide a more accurate response. There are a vast number of examples of misinformation that can be compiled and tagged. Thus, incorporating all the possible examples into the prompt increases the computing resources required to process the prompt by the LLM. The increased computing resources also often leads to increased latency in generating the response (see Ghaeini [0012]). The technology disclosed herein, among other things, provides solutions to the above problem by providing systems and methods that dynamically generate prompts based on the input content. The dynamic generation of the prompts results in prompts that are more computationally efficient while preserving the clarity and quality of the prompts. With the dynamic prompt generation, the input content is preprocessed to determine which categories and/or examples are most closely related to the input content. Based on that similarity determination, only the examples and/or categories that are determined to be most closely related (e.g., exceeding a similarity metric) are incorporated into the prompt. This allows for the examples and/or categories that are least likely to be useful for evaluation of the input content to be omitted from the prompt. Accordingly, the dynamically generated prompt allows for improved computational performance by the LLM (when the LLM processes the prompt) while still retaining the data that is most likely to lead to an accurate evaluation of the input content (see Ghaeini [0013]). Additionally, the system of Ghaeini can be used for a variety of tasks (see Ghaeini [0067]) . 07-21-aia AIA Claim 9 is rejected under 35 U.S.C. 103 as being unpatentable over Fabian in view of Mathur in further view of Eberhardt et al. (US 12346712 B1, published 07/01/2025), hereinafter Eberhardt . Regarding claim 9, Fabian in view of Mathur teaches all the limitations of claim 8, further comprising: wherein determining the environment representation of the user interface comprises determining a browser representation of the user interface (Fabian Figs 1-13; [0028], the reply from the LLM includes a specific suggestion for modifying the spreadsheet in the specific way; The LLM, upon receiving the configured prompt, generates a reply which suggests adding a column which computes a total sales quantity based on year-to-date (YTD) sales data along with a spreadsheet formula to calculate the quantity; the user may then direct the application to implement; [0032], the scope of the problem refers to the domain in which the problem is to be addressed. For example, the prompt may instruct the LLM to limit its reply to Microsoft Excel® formulas and not Google Sheets formulas. The tasks to be completed include instructions, such as, “generate a formula to do X and self-evaluate the formula.” Illustrative examples can include an example of how to self-evaluate a formula when the formula is relatively complex. An illustrative example provides the LLM with guidance on how to complete a task of the prompt; [0043], The spreadsheet environment of application service 110 may be implemented a natively installed and executed application, a browser-based application, or a mobile application; [0048-0049], Upon receiving the user's selection of the first suggestion in task pane 144, application service 110 implements the suggestion by adding a column (not shown) to the spreadsheet data; [0059], browser-based application; [0060], Application service 110 receives the input and generates a prompt to be submitted to LLM service 120 based on the input. The prompt includes contextual information, such as a chat history and spreadsheet data or metadata, such as a table name, worksheet name, or spreadsheet name, row and column headers, and at least a subset of the data, e.g., the first five or ten rows of data. Contextual information can also include the recent or latest actions performed by the user on the spreadsheet, such as the user creating a new column or entering a formula; [0084-0087], The user clicks the “Add column” button, causing one or more of application components 303 to modify data table 502 to include the new column and to fill the column according to instructions provided in the reply; [0088-0089], LLM 330 generates two formulas along with the descriptions and explanations and determines, in FIG. 5F, that the formulas are of moderate and high relevance to the input. LLM 330 transmits its reply to prompt engine 305 which, in turn, generates a response and sends the response to user interface 307 for display; see also [0094-0103], [0107-0111]) However, Fabian in view of Mathur fails to expressly disclose a hypertext markup language representation of the user interface. In the same field of endeavor, Eberhardt teaches: a hypertext markup language representation of the user interface (Eberhardt Figs. 1-10; col. 1 [line 36], generating a prompt including at least some of the component metadata, at least some of the context information, and an indication of one or more available response elements, the response elements including one or more of agent calls, data link queries, actions, or response text; providing the prompt to a large language model (LLM); receiving an output from the LLM indicating one or more response elements; col. 9 [line 32], Each of the components of a user interface, such as the example components 202-210 in FIG. 2, may be associated with component metadata, such as facts, actions, data links, and/or other categories or types of metadata. The component metadata may be included in the user interface code (e.g., as metadata in HTML code associated with the user interface) or may be accessible to the AAS 110, such as in a table or other data structure with associations between user interface components and component metadata. As discussed further herein, this component metadata may be provided to an LLM 130 as context for determining the most relevant information to provide to the user, additional information that may be obtained from an external service, or an action to perform) It would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to have incorporated a hypertext markup language representation of the user interface as suggested in Eberhard into Fabian in view of Mathur. Doing so would be desirable because embodiments of the present disclosure relate to devices, systems, and methods that employ customized artificial intelligence to assist a user of a software application (see Eberhardt col. 1 [line 6]). Software applications, such as those that are used for business applications, can display a variety of information. Often the amount of information available to a user, particularly those using a mobile device, such as a smart phone or tablet, may be overwhelming when displayed on the user device. Additional information that is available and actions that are possible with reference to the displayed information may not be known to the user. Thus, systems and methods for improving user interactions with software applications is needed (see Eberhardt col. 1 [line 18]). The techniques described herein relate to a method, wherein one or more facts includes domain-specific knowledge usable by the LLM to better response to the user input (see Eberhardt col. 3 [line 23]). Component metadata may be included in the user interface code (e.g., as metadata in HTML code associated with the user interface) or may be accessible to the AAS 110, such as in a table or other data structure with associations between user interface components and component metadata. As discussed further herein, this component metadata may be provided to an LLM 130 as context for determining the most relevant information to provide to the user (see Eberhardt col. 9 [line 32]) . Response to Arguments The Examiner acknowledges the Applicant’s amendments to claims 1-7, 10-16, and 18-20. The rejection of claims 1-6 and 8-20 under 35 U.S.C. 112(b) is respectfully withdrawn. Claim 7 stands rejected under 35 U.S.C. 112(b). Regarding independent claim 1, the Applicant alleges that Fabian as described in the previous Office action, does not explicitly teach "generating, from the lookahead prompt and using a large language model, an estimated lookahead plan describing one or more actions for performing the task, the one or more actions involving use of one or more interactive elements from the set of interactive elements" and "generating, from the estimated lookahead plan and using one or more large language models, instructions to perform a next action in a sequence for performing the task, the instructions indicating an interactive element selected from the set of interactive elements and prompting user interaction with the interactive element via the user interface in performance of the next action”, as has been amended to the claim. Examiner has therefore rejected independent claim 1 under 35 U.S.C § 103 as unpatentable over Fabian in view of Mathur. Specifically, the Applicant alleges Fabian discusses data that can be manipulated in performance of a task, not an interactive element to be interacted with in performance of the task (see remarks p. 15). As shown in Figs. 8, 9, and 10 (see also [0099-0111]), the responses include specific actions to be performed with specific displayed rows and columns within the displayed spreadsheet. Examiner notes that the claims place no limitations on what the interactive elements must comprise. Thus, Fabian’s disclosure of interactive elements is considered to teach the instructions indicating an interactive element selected from the set of interactive elements (Figs. 8, 9, and 10; [0099-0111]). Mathur is cited to clarify the instructions indicating an interactive element selected from the set of interactive elements and prompting user interaction with the interactive element via the user interface in performance of the next action (Figs. 1-7; [0015], [0021], [0024], [0036], [0044], [0045], [0051]). Applicant further alleges Fabian fails to disclose (i) generating an estimated lookahead plan describing one or more actions for performing the task and (ii) generating the instructions from the estimated lookahead plan. Examiner respectfully disagrees. As discussed in the rejection above, Fabian discloses a user can enter a natural language query which is translated to the prompt engine ([0082]). The Prompt engine 305 configures the prompt including parameters to direct LLM 330 to provide a focused response to the input ([0067]). With the prompt configured, prompt engine 305 sends the prompt to LLM 330. LLM 330 generates a reply to the prompt and transmits the reply to prompt engine 305 ([0084]). Upon receiving LLM 330's reply, prompt engine 305 post-processes the reply to generate a response for display in user interface 307. Post-processing the reply includes extracting information from the reply according to the output formatting rules, such as a description of the suggestion and instructions by which the suggestion can be implemented either by application components 303 or by the user ([0087]). Examiner notes the claims place no limitations on what the estimated lookahead plan or instructions must comprise. Thus, Fabian’s disclosure of generating, from the lookahead prompt and using a large language model, an estimated lookahead plan describing one or more actions for performing the task, the one or more actions involving use of one or more interactive elements from the set of interactive elements; and generating, from the estimated lookahead plan and using one or more large language models, instructions to perform a next action in a sequence for performing the task, the instructions indicating an interactive element selected from the set of interactive elements (Fabian Figs 1-13; [0028], [0048-0049], [0084], [0087-0089], [0094-0103], [0107-0111]) is considered to teach the claim limitations. Similar arguments have been presented for claims 10 and 18 and thus, Applicant’s arguments are not persuasive for the same reasons. Applicant states that the dependent claims recite all the limitations of the independent claims, and thus, are allowable in view of the remarks set forth regarding the independent claims. However, as discussed above, Fabian in view of Mathur is considered to teach the independent claims, and consequently, the dependent claims are rejected. Applicant further alleges that the currently amended independent claims are directed to patent-eligible subject matter and not directed to a mental process or any other abstract idea. Even assuming, arguendo, that the claims are directed to an abstract idea, the amendments to the claims integrate any such abstract idea into a practical application (see remarks pp. 19-20). Examiner respectfully disagrees. As discussed in the rejection above, the claim is directed to an abstract idea that encompasses mental processes including evaluations or observations that are practically capable of being performed in the human mind with the assistance of pen and paper, and mathematical concepts that are achievable through mathematical computation. The claim places no limits on how the generating and determining are performed. That is, nothing in the claim element precludes the step from practically being performed in the mind. As further discussed above, when viewed in combination, the additional elements in the claims do not integrate the recited judicial exception into a practical application (Step 2A, Prong Two: NO), and the additional elements individually or in combination with the judicial exception do not provide an inventive concept; so, the claims as a whole do not amount to significantly more than the abstract idea. (Step 2B: NO). The claims are not eligible. Conclusion 07-96 AIA The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. Dotan-Cohen; (US 20250225008 A1) see Figs. 1-10 and [0107-0114] . Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL . See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a). A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action. Any inquiry concerning this communication or earlier communications from the examiner should be directed to JOHN T REPSHER III whose telephone number is (571)272-7487. The examiner can normally be reached Monday - Friday, 8AM-5PM EST. Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice. If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Jennifer Welch can be reached at (571) 272-7212. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300. Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000. /JOHN T REPSHER III/ Primary Examiner, Art Unit 2143 Application/Control Number: 18/670,398 Page 2 Art Unit: 2143 Application/Control Number: 18/670,398 Page 3 Art Unit: 2143 Application/Control Number: 18/670,398 Page 4 Art Unit: 2143 Application/Control Number: 18/670,398 Page 5 Art Unit: 2143 Application/Control Number: 18/670,398 Page 6 Art Unit: 2143 Application/Control Number: 18/670,398 Page 7 Art Unit: 2143 Application/Control Number: 18/670,398 Page 8 Art Unit: 2143 Application/Control Number: 18/670,398 Page 9 Art Unit: 2143 Application/Control Number: 18/670,398 Page 10 Art Unit: 2143 Application/Control Number: 18/670,398 Page 11 Art Unit: 2143 Application/Control Number: 18/670,398 Page 12 Art Unit: 2143 Application/Control Number: 18/670,398 Page 13 Art Unit: 2143 Application/Control Number: 18/670,398 Page 14 Art Unit: 2143 Application/Control Number: 18/670,398 Page 15 Art Unit: 2143 Application/Control Number: 18/670,398 Page 16 Art Unit: 2143 Application/Control Number: 18/670,398 Page 17 Art Unit: 2143 Application/Control Number: 18/670,398 Page 18 Art Unit: 2143 Application/Control Number: 18/670,398 Page 19 Art Unit: 2143 Application/Control Number: 18/670,398 Page 20 Art Unit: 2143 Application/Control Number: 18/670,398 Page 21 Art Unit: 2143 Application/Control Number: 18/670,398 Page 22 Art Unit: 2143 Application/Control Number: 18/670,398 Page 23 Art Unit: 2143 Application/Control Number: 18/670,398 Page 24 Art Unit: 2143 Application/Control Number: 18/670,398 Page 25 Art Unit: 2143 Application/Control Number: 18/670,398 Page 26 Art Unit: 2143 Application/Control Number: 18/670,398 Page 27 Art Unit: 2143 Application/Control Number: 18/670,398 Page 28 Art Unit: 2143 Application/Control Number: 18/670,398 Page 29 Art Unit: 2143 Application/Control Number: 18/670,398 Page 30 Art Unit: 2143 Application/Control Number: 18/670,398 Page 31 Art Unit: 2143 Application/Control Number: 18/670,398 Page 32 Art Unit: 2143 Application/Control Number: 18/670,398 Page 33 Art Unit: 2143 Application/Control Number: 18/670,398 Page 34 Art Unit: 2143 Application/Control Number: 18/670,398 Page 35 Art Unit: 2143 Application/Control Number: 18/670,398 Page 36 Art Unit: 2143 Application/Control Number: 18/670,398 Page 37 Art Unit: 2143 Application/Control Number: 18/670,398 Page 38 Art Unit: 2143 Application/Control Number: 18/670,398 Page 39 Art Unit: 2143 Application/Control Number: 18/670,398 Page 40 Art Unit: 2143 Application/Control Number: 18/670,398 Page 41 Art Unit: 2143 Application/Control Number: 18/670,398 Page 42 Art Unit: 2143 Application/Control Number: 18/670,398 Page 43 Art Unit: 2143 Application/Control Number: 18/670,398 Page 44 Art Unit: 2143 Application/Control Number: 18/670,398 Page 45 Art Unit: 2143 Application/Control Number: 18/670,398 Page 46 Art Unit: 2143 Application/Control Number: 18/670,398 Page 47 Art Unit: 2143 Application/Control Number: 18/670,398 Page 48 Art Unit: 2143 Application/Control Number: 18/670,398 Page 49 Art Unit: 2143 Application/Control Number: 18/670,398 Page 50 Art Unit: 2143 Application/Control Number: 18/670,398 Page 51 Art Unit: 2143 Application/Control Number: 18/670,398 Page 52 Art Unit: 2143 Application/Control Number: 18/670,398 Page 53 Art Unit: 2143 Application/Control Number: 18/670,398 Page 54 Art Unit: 2143 Application/Control Number: 18/670,398 Page 55 Art Unit: 2143 Application/Control Number: 18/670,398 Page 56 Art Unit: 2143 Application/Control Number: 18/670,398 Page 57 Art Unit: 2143 Application/Control Number: 18/670,398 Page 58 Art Unit: 2143 Application/Control Number: 18/670,398 Page 59 Art Unit: 2143 Application/Control Number: 18/670,398 Page 60 Art Unit: 2143 Application/Control Number: 18/670,398 Page 61 Art Unit: 2143 Application/Control Number: 18/670,398 Page 62 Art Unit: 2143 Application/Control Number: 18/670,398 Page 63 Art Unit: 2143 Application/Control Number: 18/670,398 Page 64 Art Unit: 2143 Application/Control Number: 18/670,398 Page 65 Art Unit: 2143 Application/Control Number: 18/670,398 Page 66 Art Unit: 2143 Application/Control Number: 18/670,398 Page 67 Art Unit: 2143 Application/Control Number: 18/670,398 Page 68 Art Unit: 2143 Application/Control Number: 18/670,398 Page 69 Art Unit: 2143 Application/Control Number: 18/670,398 Page 70 Art Unit: 2143 Application/Control Number: 18/670,398 Page 71 Art Unit: 2143 Application/Control Number: 18/670,398 Page 72 Art Unit: 2143 Application/Control Number: 18/670,398 Page 73 Art Unit: 2143 Application/Control Number: 18/670,398 Page 74 Art Unit: 2143 Application/Control Number: 18/670,398 Page 75 Art Unit: 2143 Application/Control Number: 18/670,398 Page 76 Art Unit: 2143 Application/Control Number: 18/670,398 Page 77 Art Unit: 2143 Application/Control Number: 18/670,398 Page 78 Art Unit: 2143 Application/Control Number: 18/670,398 Page 79 Art Unit: 2143 Application/Control Number: 18/670,398 Page 80 Art Unit: 2143 Application/Control Number: 18/670,398 Page 81 Art Unit: 2143 Application/Control Number: 18/670,398 Page 82 Art Unit: 2143 Application/Control Number: 18/670,398 Page 83 Art Unit: 2143 Application/Control Number: 18/670,398 Page 84 Art Unit: 2143 Application/Control Number: 18/670,398 Page 85 Art Unit: 2143 Application/Control Number: 18/670,398 Page 86 Art Unit: 2143 Application/Control Number: 18/670,398 Page 87 Art Unit: 2143 Application/Control Number: 18/670,398 Page 88 Art Unit: 2143 Application/Control Number: 18/670,398 Page 89 Art Unit: 2143 Application/Control Number: 18/670,398 Page 90 Art Unit: 2143 Application/Control Number: 18/670,398 Page 91 Art Unit: 2143
Prosecution Insights