All Services

Reorders recognised sections in a training lesson document according to a standard pedagogical template, using deterministic processing.

Applies a structured section template to supplied Markdown or plain text to produce a formatted training lesson document

Normalises whitespace in training lesson Markdown or plain text documents up to 20,000 characters using deterministic processing.

Computes confidence calibration bins for a text classifier that categorizes training questions into concept, procedure, example, or assessment classes.

Audits up to 200 labelled training question text records to detect exact normalised duplicates within the dataset.

Generates a fixed-holdout learning curve analysis for a naive-Bayes text classifier trained to categorize training content as concept, procedure, example, or assessment.

Splits a labelled dataset of training questions into grouped train/test subsets using a naive-Bayes baseline approach, returning bounded split deliverables for up to 200 records.

Identifies low-confidence predictions in a labeled training dataset of educational text records classified as concept, procedure, example, or assessment, returning a review queue for human inspection.

Computes macro, micro, and weighted aggregate classification scores (precision, recall, F1) for a set of labelled training question records with actual vs. predicted categories.

Computes prediction error records for a naive-Bayes text classifier trained on training question categories (concept, procedure, example, assessment) given up to 200 labelled records with actual vs predicted labels and class probabilities.

Detects text overlap between labelled training records and evaluation texts for ML training question classification datasets.

Trains a local naive-Bayes text classifier on up to 200 labelled records and classifies query texts into one of four educational content categories.

Verifies and audits Base blockchain transactions by matching on-chain payment records against expected recipients, assets, and amounts.

Retrieves deterministic on-chain evidence for a Base blockchain transaction by hash, suitable for agent audit trails.

Analyzes supplied transaction fee observation records by computing category frequencies and a date-by-category pivot table for a chosen numeric field (fee_units or transaction_count).

Analyzes up to 200 transaction fee observation records to produce two local analytical reports: a duplicate key report and category frequency distribution.

Analyzes up to 200 transaction fee observation records to produce a duplicate key report and a numeric interval filter report for fee_units or transaction_count fields.

Runs two local analytical reports on supplied transaction fee observation rows: a duplicate key detection report and a secondary-measure weighted mean calculation.

Deduplicates transaction fee observation rows and computes a Base-100 index series for fee_units or transaction_count fields across up to 200 supplied records.

Deduplicates transaction fee observation rows and computes a trailing mean series for fee_units or transaction_count fields, returning two analytical reports for up to 200 rows.

Computes grouped numeric summaries and fills missing values in transaction fee observation records (fee_units, transaction_count) using local analytical engines.

Analyzes up to 200 rows of transaction fee observation records by producing a grouped numeric summary and a two-measure correlation report for fee_units and transaction_count.

Computes a numeric distribution summary and an equal-width histogram for user-supplied transaction fee observation records (fee_units or transaction_count).

Applies a numeric interval filter and computes adjacent row changes on user-supplied transaction fee observation records (fee_units, transaction_count), returning two analytical reports for up to 200 rows.

Applies a numeric interval filter and computes a Base-100 index series on supplied transaction fee observation records (fee_units, transaction_count), returning up to 2 analytical reports for up to 200 rows.

Applies numeric interval filtering and keyed dataset comparison to supplied transaction fee observation records (fee_units, transaction_count), producing 2 analytical reports for up to 200 rows.

Filters transaction fee observation rows by a numeric interval and computes a trailing mean series, returning two analytical reports for up to 200 rows of fee data.

Applies numeric row ordering and a base-100 index series to up to 200 transaction fee observation records, producing two analytical reports on fee_units or transaction_count data.

Ranks transaction fee observation rows numerically and computes correlation between fee_units and transaction_count for up to 200 supplied records.

Audits data quality and computes adjacent row changes for transaction fee observation records (fee_units, transaction_count) across up to 200 rows.

Runs two local analytical reports on supplied transaction fee observation rows: a record quality audit and a base-100 index series, for up to 200 rows per call.

Runs two local analytical reports on up to 200 transaction fee observation rows: a record quality audit and a date-by-category pivot on either fee_units or transaction_count.

Audits data quality and fills missing values (via zero, mean, or median) for transaction fee observation records containing fee_units and transaction_count fields.

Runs a record quality audit and keyed dataset comparison on up to 200 rows of transaction fee observations (fee_units, transaction_count) and returns two independent analytical reports.

Audits data quality and computes numeric distribution statistics for transaction fee observation records (fee_units, transaction_count) — up to 200 rows, two reports per call.

Runs two local analytical reports on supplied transaction fee observation rows: a record quality audit and a numeric interval filter on fee_units or transaction_count.

Audits record quality and computes a secondary-measure weighted mean for up to 200 transaction fee observation rows (fee_units, transaction_count).

Runs three local analytical reports (record quality audit, string normalisation, and adjacent row changes) on up to 200 supplied transaction fee observation rows containing fee_units and transaction_count fields.

Audits record quality, normalises strings, and filters by numeric interval for up to 200 transaction fee observation rows, returning 3 independent analytical reports.

Computes a transaction-count-weighted mean and fills missing values in supplied transaction fee observation records using a chosen imputation method.

Applies string normalisation and adjacent row change analysis to up to 200 transaction fee observation records, returning two local analytical reports on fee_units or transaction_count fields.

Runs string normalisation and numeric distribution analysis on up to 200 rows of transaction fee observation records (fee_units, transaction_count), returning two independent analytical reports.

Normalises string fields and filters transaction fee observation rows by a numeric interval across fee_units or transaction_count, producing 2 analytical reports for up to 200 rows.

Runs two local analytical reports on uploaded transaction fee data: a Pearson/Spearman-style correlation between fee_units and transaction_count, and a date-by-instrument-group pivot table, for up to 200 rows at $0.01 USDC.

Computes Pearson/Spearman correlation between fee_units and transaction_count, and fills missing values using zero, mean, or median imputation, for up to 200 rows of transaction fee data.

Runs two local statistical reports on user-supplied transaction fee records: a correlation analysis between fee_units and transaction_count, plus a keyed comparison between two datasets.

Computes a statistical distribution summary of the ratio of transaction fees to total mining reward (fee + block subsidy) across a set of provided block-level data points.

Computes the local finite-change elasticity of transaction fees as a share of total Bitcoin mining reward, given user-supplied baseline fee and subsidy values and a percentage step size.

Computes a rolling mean of Bitcoin transaction fee share in mining rewards over a sliding window across a series of block data rows.

Calculates the gap between a target value and the combined transaction fee plus block subsidy in a Bitcoin mining reward scenario, given user-supplied mining assumptions.

Computes a weighted arithmetic mean of the transaction-fee share in Bitcoin block rewards across up to 200 supplied mining data rows.

Computes adjacent row changes and secondary-measure weighted mean for transcript preparation records (transcribed_minutes, source_minutes) across up to 200 rows, returning two analytical reports.

Computes category-level frequency counts and filters transcript preparation records by a numeric interval across transcribed_minutes or source_minutes fields, returning two analytical reports for up to 200 rows.

Computes category frequencies and a secondary-measure weighted mean for up to 200 transcript preparation records (transcribed_minutes, source_minutes) in two independent local analytical reports.

Analyzes up to 200 transcript preparation records by producing a daily sum time series and a keyed dataset comparison report for transcribed_minutes and source_minutes fields.

Analyzes transcript preparation records to produce two local analytical reports: a duplicate key report and an adjacent row change report for transcribed_minutes or source_minutes.

Analyzes up to 200 transcript preparation records to produce a duplicate key report and a grouped numeric summary for transcribed_minutes or source_minutes fields.

Analyzes up to 200 transcript preparation records to produce a duplicate key report and numeric distribution for transcribed_minutes or source_minutes fields.

Analyzes up to 200 transcript preparation records to produce a duplicate key report and a numeric interval filter report for transcribed_minutes or source_minutes fields.

Analyzes up to 200 transcript preparation records (transcribed_minutes, source_minutes) and returns two reports: a duplicate key report and a numeric row ordering report.

Generates an equal-width histogram and fills missing values in transcript preparation records (transcribed_minutes, source_minutes) for up to 200 rows.

Analyzes transcript preparation records by generating an equal-width histogram for a selected measure and computing the Pearson/linear correlation between transcribed_minutes and source_minutes across up to 200 rows.

Performs exact row deduplication and Base-100 index series analysis on up to 200 transcript preparation records containing transcribed_minutes and source_minutes fields.

Deduplicates transcript preparation records by exact row matching and compares two keyed datasets (transcribed_minutes vs source_minutes) to produce two independent analytical reports.

Deduplicates transcript preparation records by exact row matching and computes numeric distribution statistics for transcribed_minutes or source_minutes fields across up to 200 rows.

Deduplicates transcript preparation records by exact row matching and computes a secondary-measure weighted mean for transcribed_minutes or source_minutes fields across up to 200 rows.

Analyzes transcript preparation records (transcribed_minutes, source_minutes) to produce two local analytical reports: a grouped numeric summary and an adjacent-row change analysis, for up to 200 rows.

Computes a grouped numeric summary and fills missing values for transcript preparation records (transcribed_minutes, source_minutes) across up to 200 rows, returning two analytical reports.

Analyzes transcript preparation records by producing a grouped numeric summary and a keyed dataset comparison for up to 200 rows of transcribed_minutes and source_minutes data.

Computes grouped numeric statistics and a trailing mean time series for transcript preparation records (transcribed_minutes, source_minutes) across up to 200 rows.

Computes a grouped numeric summary and a two-measure correlation report for transcript preparation records (transcribed_minutes, source_minutes) from up to 200 supplied rows.

Analyzes up to 200 transcript preparation records to produce two reports: a numeric distribution analysis and an adjacent row changes report for transcribed_minutes or source_minutes.

Runs two local analytical reports on transcript preparation records: a numeric distribution analysis and a date-by-category pivot, for up to 200 rows of transcribed_minutes and source_minutes data.

Computes numeric distribution statistics and fills missing values (zero, mean, or median) for up to 200 rows of transcript timing data (transcribed_minutes, source_minutes), returning two analytical reports.

Analyzes transcript preparation records by computing a numeric distribution and filtering rows by a numeric interval for transcribed_minutes or source_minutes fields, across up to 200 rows.

Computes a numeric distribution report and a numeric row ordering report for transcript preparation records (transcribed_minutes, source_minutes) across up to 200 rows.

Analyzes transcript preparation records by computing a numeric distribution and a trailing mean time series for up to 200 rows of transcribed_minutes or source_minutes data.

Filters transcript preparation rows by a numeric interval on a chosen field and computes adjacent row changes, returning two analytical reports for up to 200 rows.

Filters transcript preparation records by a numeric interval on a chosen field and computes a Base-100 indexed series, returning 2 analytical reports for up to 200 rows.

Filters transcript minute records to a numeric interval and fills missing values using zero, mean, or median imputation, returning two analytical reports for up to 200 rows.

Sorts transcript preparation rows numerically and fills missing transcribed_minutes or source_minutes values using zero, mean, or median imputation — up to 200 rows, two reports per call.

Runs two analytical reports on transcript preparation records: a record quality audit and a date-by-category pivot, for up to 200 rows of transcribed/source minutes data.

Runs a record quality audit and an equal-width histogram analysis on up to 200 transcript preparation rows (transcribed_minutes, source_minutes) and returns both reports.

Runs three local analytical reports on transcript preparation records: a record quality audit, exact row deduplication, and numeric distribution analysis for transcribed_minutes or source_minutes.

Performs a data quality audit and grouped numeric summary on up to 200 transcript preparation records containing transcribed and source minute fields.

Audits data quality and computes a secondary-measure weighted mean for up to 200 transcript preparation records containing transcribed_minutes and source_minutes fields.

Runs three local analytical reports on transcript preparation records: record quality audit, string normalisation, and base-100 index series — for up to 200 rows of transcribed_minutes and source_minutes data.

Runs three local analytical reports on transcript preparation data (transcribed_minutes, source_minutes): record quality audit, string normalisation, and daily sum series — up to 200 rows per call.

Runs three local analytical reports on transcript preparation records — a data quality audit, string normalisation, and grouped numeric summary — for up to 200 rows of transcribed_minutes and source_minutes data.

Computes a secondary-measure weighted mean and performs a keyed dataset comparison on up to 200 transcript preparation records (transcribed_minutes, source_minutes), returning two analytical reports.

Normalises string fields and fills missing values in transcript preparation records containing transcribed_minutes and source_minutes, producing two analytical reports for up to 200 rows.

Normalises string fields and filters rows by a numeric interval (min/max) for transcript preparation records containing transcribed_minutes and source_minutes, returning two analytical reports for up to 200 rows.

Normalises transcribed and source minute strings and computes two-measure correlation across up to 200 transcript preparation rows, returning two analytical reports.

Runs two analytical reports on transcript preparation data: a trailing mean time series and a date-by-category pivot table, for up to 200 rows of transcribed/source minutes records.

Computes Pearson correlation between transcribed_minutes and source_minutes, and fills missing values in a chosen field using zero, mean, or median imputation — across up to 200 transcript preparation rows.

Computes Pearson/statistical correlation between transcribed_minutes and source_minutes, plus a keyed comparison between two supplied datasets, across up to 200 transcript preparation rows.

Validates cell length constraints in a CSV table of transfer records (up to 1000 rows), reporting any cells that exceed or violate configured length limits.

Runs two analytical reports on transformation throughput data: adjacent row-to-row changes and an equal-width histogram, using local analytical engines on up to 200 supplied records.

Analyzes transformation throughput records by computing adjacent row changes and filling missing values using a chosen strategy (zero, mean, or median) for up to 200 rows.

Analyzes transformation throughput records by computing adjacent row changes and IQR-based outlier flags for processed_rows or duration_seconds fields across up to 200 rows.