All Services

Runs three data-quality reports on supplier qualification records: a record quality audit, string normalisation, and explicit missing-value fill for checked_criteria and met_criteria fields — up to 200 rows per call.

Runs three analytical reports on supplier qualification data: a record quality audit, string normalisation, and numeric distribution analysis for up to 200 rows.

Runs three local analytical reports on supplier qualification row data: a record quality audit, string normalisation, and correlation between checked_criteria and met_criteria fields.

Computes a secondary-measure weighted mean and a date-by-category pivot table for up to 200 supplier qualification records in a single paid API call.

Computes a secondary-measure weighted mean and fills missing values for supplier qualification records (checked_criteria, met_criteria) across up to 200 rows, returning 2 analytical reports.

Computes a secondary-measure weighted mean and performs a keyed dataset comparison across up to 200 supplier qualification records (checked_criteria, met_criteria).

Analyzes up to 200 supplier qualification rows by performing string normalization and detecting adjacent row changes on checked_criteria or met_criteria fields, returning two analytical reports.

Normalises supplier qualification strings and computes Base-100 index series for checked_criteria and met_criteria fields across up to 200 rows, returning two analytical reports.

Normalises strings and deduplicates exact rows in supplier qualification records (checked_criteria, met_criteria) for up to 200 rows, returning 2 analytical reports.

Normalises string fields and fills missing values in supplier qualification records (checked_criteria, met_criteria) for up to 200 rows, returning 2 analytical reports per call.

Normalises strings and compares two supplier qualification datasets (checked_criteria vs met_criteria) row-by-row, producing 2 analytical reports for up to 200 rows each.

Normalises company_group strings and sorts supplier qualification rows numerically by checked_criteria or met_criteria, returning two analytical reports for up to 200 rows.

Normalises string fields and computes a secondary-measure weighted mean for supplier qualification records, returning two analytical reports for up to 200 rows.

Normalises string fields and computes correlation between checked_criteria and met_criteria across up to 200 supplier qualification rows, producing two independent analytical reports.

Computes a trailing mean time series and a keyed dataset comparison for supplier qualification records (checked_criteria, met_criteria) across up to 200 rows, returning two analytical reports.

Computes a trailing moving average series and a secondary-measure weighted mean for supplier qualification data (checked_criteria, met_criteria) across up to 200 rows, returning two analytical reports.

Runs two local analytical reports on supplier qualification data: a Pearson correlation between checked_criteria and met_criteria, and a date-by-category pivot table, for up to 200 rows.

Computes correlation between checked_criteria and met_criteria for supplier qualification records and fills missing values using a chosen method (zero, mean, or median)

Searches live public web pages for supplier verification data (business registries, sanctions lists, etc.) relevant to agent builders and returns source links with snippets.

Searches live public web pages for supplier verification data, returning source links and snippets relevant to developer-focused queries about business registries and sanctions databases.

Searches live public web pages for supplier verification data (business registries, sanctions lists, etc.) relevant to operations teams, returning source links and snippets.

Scans a support article in Markdown or plain text and identifies duplicate paragraphs within the document.

Audits the heading hierarchy of a support article document supplied as Markdown or plain text, checking for structural correctness and proper nesting.

Counts and inventories number tokens (digits, dates, quantities, etc.) found in a support article document supplied as Markdown or plain text.

Scans a support article (Markdown or plain text) and inventories all quoted passages found within it, up to 20,000 characters.

Normalises whitespace in support article Markdown or plain text, up to 20,000 characters, using deterministic processing

Computes confidence calibration bins for a support-intent classifier across billing, technical, account, and other categories using up to 200 labelled text records.

Audits up to 200 labelled support-intent text records for exact normalised duplicates across billing, technical, account, and other categories.

Generates a fixed-holdout learning curve for a naive-Bayes support intent classifier (billing/technical/account/other) using up to 200 labelled text records.

Splits a labelled support-intent dataset (billing/technical/account/other) into grouped train/test partitions and returns dataset split deliverables with naive-Bayes baseline evaluation metrics.

Computes label distribution statistics for a set of labelled support intent text records across billing, technical, account, and other categories.

Exports up to 200 labelled customer support text records as a JSONL file, with each record tagged by support intent category (billing, technical, account, or other).

Accepts up to 200 labelled support-intent text records with actual/predicted class pairs and probabilities, and returns prediction error analysis for a naive-Bayes support-intent classifier across billing, technical, account, and other categories.

Generates a taxonomy prompt template for classifying support text records into billing, technical, account, or other intents using up to 200 labelled examples.

Validates a CSV of support ticket records and reports blank/empty rows, using configurable delimiter, cell length, and category constraints.

Validates and normalizes a CSV table of support ticket import data, trimming whitespace and converting cell values to their inferred types, for up to 1000 rows.

Computes adjacent row-to-row changes for support workload metrics (handling_minutes or resolved_count) across up to 200 typed records

Filters support workload records (by queue, date, handling minutes, or resolved count) to only those rows whose selected numeric field falls within a specified minimum–maximum range.

Sorts up to 200 support workload records numerically by handling_minutes or resolved_count, returning rows in ordered sequence.

Sorts a list of numbers or strings in ascending or descending order, optimized for survey data workflows.

Parses a JSON string and returns its root type, top-level keys, and item count, optimized for survey workflow data inspection.

Compares two string arrays and returns their unique and shared (intersection/difference) values, optimized for survey workflows.

Computes adjacent-row change analysis and fills missing values in survey participation data (respondents and invitations) for up to 200 rows, returning two analytical reports.

Computes adjacent row-to-row changes and secondary-measure weighted mean for survey participation data (respondents and invitations) over up to 200 rows, returning two analytical reports.

Computes a base-100 indexed time series and fills missing values for survey participation data (respondents and/or invitations) across up to 200 rows.

Computes category frequency distributions and fills missing values in survey participation data (respondents/invitations) for up to 200 rows, returning two analytical reports.

Analyzes survey participation data (respondents and invitations) to produce category frequency counts by course group and numeric row ordering, delivering two analytical reports for up to 200 rows.

Computes category frequencies and secondary-measure weighted mean from survey participation records (respondents/invitations) for up to 200 rows, returning two analytical reports.

Runs two local analytical reports on supplied survey participation data: identifies duplicate row keys and filters rows where a chosen numeric field falls within a specified range.

Generates two analytical reports from survey participation data: an equal-width histogram of respondents or invitations, and a date-by-category pivot table, for up to 200 rows.

Deduplicates survey participation records and computes a Base-100 index series for respondents or invitations across up to 200 rows, delivering two analytical reports.

Deduplicates survey participation records and computes category-level frequency distributions across course groups, returning two analytical reports for up to 200 rows.

Deduplicates exact duplicate rows and computes numeric distribution statistics for survey participation data (respondents or invitations) across up to 200 rows.

Deduplicates survey participation rows exactly and computes a secondary-measure weighted mean for respondent/invitation data, returning two analytical reports for up to 200 rows.

Deduplicates survey participation records by exact row and computes a trailing mean time series for respondent or invitation counts across up to 200 rows.

Analyzes survey participation data (respondents and invitations) by computing grouped numeric summaries and adjacent row changes across up to 200 rows.

Computes grouped numeric summaries and a secondary-measure weighted mean for survey participation data (respondents and invitations) across up to 200 rows, returning two independent analytical reports.

Computes numeric distribution and secondary-measure weighted mean for survey participation data (respondents and invitations) across up to 200 rows, returning two analytical reports.

Filters survey participation data (respondents or invitations) by a numeric range and computes a Base-100 index series, delivering two analytical reports for up to 200 rows of supplied data.

Filters survey participation rows by a numeric range on a chosen measure and computes a two-measure correlation (respondents vs. invitations) across up to 200 supplied rows, returning two independent analytical reports.

Analyzes survey participation data (respondents and invitations) by computing numeric row ordering and detecting adjacent row changes across up to 200 rows.

Computes numeric row ordering and base-100 index series for survey participation data (respondents or invitations) across up to 200 rows with 2 analytical reports.

Ranks survey participation rows numerically and computes a daily sum time series for respondents or invitations data, returning two analytical reports for up to 200 rows.

Analyzes survey participation data (respondents and invitations) by producing two reports: a numeric row ordering and a date-by-category pivot, for up to 200 rows.

Sorts survey participation rows numerically and fills missing respondent/invitation values using zero, mean, or median imputation across up to 200 records.

Sorts survey participation records by a numeric field and filters rows to a specified numeric interval, returning up to 2 analytical reports for up to 200 rows.

Analyzes survey participation data (respondents/invitations) by computing numeric row ordering and a trailing mean series for up to 200 supplied rows.

Runs three local analytical reports on survey participation data — record quality audit, exact row deduplication, and numeric interval filtering — for up to 200 rows of respondent/invitation records.

Runs three local analytical reports on survey participation data (respondents/invitations): record quality audit, exact row deduplication, and numeric row ordering — up to 200 rows per call.

Audits data quality and filters by numeric interval for survey participation records (respondents and invitations) across up to 200 rows, returning two analytical reports.

Audits data quality and ranks rows numerically for survey participation records (respondents and invitations) across up to 200 rows, delivering two independent analytical reports.

Performs a data quality audit and weighted mean calculation on survey participation records (respondents and invitations) for up to 200 rows, returning two independent analytical reports.

Runs three local analytical reports on survey participation records: record quality audit, string normalisation, and exact row deduplication — up to 200 rows, no external lookups.

Runs three data-quality reports on survey participation records (respondents/invitations): record quality audit, string normalisation, and explicit missing-value fill using zero/mean/median imputation — up to 200 rows per call.

Runs three analytical reports on survey participation records (respondents and invitations): a record quality audit, string normalisation, and keyed dataset comparison between two supplied datasets.

Runs three local analytical reports on survey participation records: record quality audit, string normalisation, and numeric row ordering — up to 200 rows per call.

Runs three analytical reports on survey participation records: a record quality audit, string normalisation, and a secondary-measure weighted mean — all processed locally against supplied data.

Runs three local analytical reports on survey participation data: record quality audit, string normalisation, and trailing mean series for respondents or invitations across up to 200 rows.

Computes a secondary-measure weighted mean and performs a keyed dataset comparison on up to 200 survey participation records (respondents/invitations)

Runs two local analytical reports on survey participation data: string normalisation and adjacent row changes, for up to 200 rows of respondent/invitation records.

Normalises string fields and computes a base-100 index series for up to 200 survey participation records (respondents or invitations) across course groups and dates.

Normalises course_group/date strings and computes daily sum time series for respondents or invitations from supplied survey participation rows, returning two analytical reports.

Runs string normalisation and equal-width histogram analysis on survey participation records (respondents or invitations) for up to 200 rows, returning two analytical reports.

Normalises string fields and produces grouped numeric summaries from survey participation records (respondents and invitations) for up to 200 rows, returning two analytical reports.

Normalises strings and compares two keyed survey participation datasets (respondents and invitations) across up to 200 rows, producing two analytical reports.

Normalises string fields and computes numeric distribution statistics for survey participation data (respondents and invitations) across up to 200 rows, returning two analytical reports.

Normalises course_group strings and filters survey participation records (respondents or invitations) by a numeric min/max interval, returning two analytical reports for up to 200 rows.

Normalises string fields and computes two-measure correlation between respondents and invitations for up to 200 survey participation records, producing two analytical reports.

Computes a trailing moving average series and a date-by-category pivot table for supplied survey participation data (respondents or invitations), returning two analytical reports for up to 200 rows.

Computes Pearson correlation between respondents and invitations, and fills missing values using zero/mean/median imputation, for up to 200 survey participation records.

Computes two analytical reports on survey response weight data: adjacent row-to-row change analysis and an equal-width histogram, for up to 200 rows.

Analyzes survey response weight records to detect adjacent row changes and fill explicit missing values, producing two analytical reports for up to 200 rows.

Computes adjacent row changes and secondary-measure weighted mean for survey response weight records, delivering two analytical reports on supplied data up to 200 rows.

Computes a Base-100 index series and fills missing values for supplied survey response weight records (response_value, sample_weight) across up to 200 rows.

Runs two local analytical reports on survey response weight data: a daily sum time series and a keyed row-by-row comparison between two supplied datasets, for up to 200 rows.

Analyzes survey response weight records to produce two local reports: a duplicate key detection report and an adjacent row changes report, for up to 200 rows.

Analyzes up to 200 survey response weight records to produce a duplicate key report and a daily sum time series for a chosen numeric field.

Analyzes up to 200 survey response weight records to produce a duplicate key report and a numeric row ordering report for the response_value or sample_weight field.

Analyzes survey response weight records to produce a duplicate key report and a trailing mean series for up to 200 rows.

Deduplicates exact duplicate rows and pivots survey response weight data by date and category, producing two analytical reports from up to 200 supplied records.