All Services

Computes the label distribution (class balance) across up to 200 labelled project-status text records, reporting counts for each status category.

Identifies and queues low-confidence predictions from a naive-Bayes project status classifier for human review, returning a bounded deliverable for up to 200 labelled text records.

Accepts up to 200 labelled text records with actual and predicted project status classifications and returns prediction error analysis using naive-Bayes evaluation tools.

Converts promotion observation data (discount_amount, units_sold) into a Base-100 indexed time series and fills missing values using zero, mean, or median imputation, returning both reports for up to 200 rows.

Analyzes promotion observation records (discount_amount, units_sold) to produce category frequency counts and numeric row ordering reports for up to 200 rows.

Analyzes up to 200 promotion observation records (discount_amount, units_sold) to produce a duplicate row_id report and category frequency counts for store_group and date fields.

Analyzes up to 200 promotion observation rows to produce a duplicate key report and fill missing values in discount_amount or units_sold fields using zero, mean, or median imputation.

Deduplicates promotion observation rows exactly and computes a secondary-measure weighted mean for discount_amount or units_sold data, returning two analytical reports from up to 200 rows.

Compares two supplied datasets of promotion observations (discount_amount, units_sold) by row key and fills missing values, producing 2 analytical reports for up to 200 rows.

Filters promotion records by a numeric interval on a selected field and computes the Pearson correlation between discount_amount and units_sold across up to 200 supplied rows.

Runs two local analytical reports on promotion data rows: numeric row ordering and daily sum series for discount_amount or units_sold fields, covering up to 200 rows.

Ranks promotion observation rows numerically and compares two keyed promotion datasets across discount_amount and units_sold fields, returning two analytical reports for up to 200 rows.

Runs two local analytical reports on promotion observation data (discount_amount, units_sold): a record quality audit and a date-by-category pivot table, for up to 200 rows.

Runs three local analytical reports on promotion observation data (discount_amount, units_sold): a record quality audit, exact row deduplication, and an equal-width histogram.

Runs three local analytical reports (record quality audit, string normalisation, and adjacent row changes) on up to 200 promotion observation records containing discount_amount and units_sold fields.

Runs two local analytical reports on promotion observation records: string normalisation of text fields and a duplicate key detection report, covering up to 200 rows of discount_amount and units_sold data.

Normalises strings and fills missing values (zero, mean, or median) for promotion observation records containing discount_amount and units_sold, producing 2 analytical reports for up to 200 rows.

Normalises string fields and computes grouped numeric summaries for promotion observation records containing discount_amount and units_sold, returning two analytical reports.

Normalises string fields and performs keyed dataset comparison across two sets of promotion observation records (discount_amount, units_sold) for up to 200 rows, producing 2 independent analytical reports.

Normalises strings and computes correlation between discount_amount and units_sold across up to 200 promotion observation rows, returning two analytical reports.

Execute a multi-step prompt chain with state tracking and optional final synthesis across sequential LLM calls

Validates cell length constraints in a property listing CSV, reporting any cells that exceed the configured maximum character length.

Validates a CSV of property listing records for whitespace anomalies, checking up to 1000 rows with configurable delimiter, cell-length, numeric range, and category allowlists.

Runs two statistical reports (adjacent row changes and equal-width histogram) on proposal preparation data (completed_sections, planned_sections) for up to 200 rows.

Analyzes proposal preparation records (completed_sections, planned_sections) to produce two local analytical reports: category frequency counts and adjacent row change calculations, for up to 200 rows.

Computes category frequency counts and a trailing moving average series for proposal preparation records (completed_sections, planned_sections) across up to 200 rows.

Computes a daily sum time series and a two-measure correlation analysis for proposal preparation data (completed_sections and planned_sections) across up to 200 rows.

Pivots proposal preparation data by date and document group, then fills missing values in completed_sections or planned_sections using zero, mean, or median imputation.

Runs two analytical reports on proposal preparation data: a date-by-category pivot table and a keyed dataset comparison between two supplied datasets (completed_sections vs planned_sections), for up to 200 rows.

Analyzes proposal preparation records to produce a duplicate key report and category frequency breakdown for up to 200 rows using local analytical engines.

Generates two local analytical reports — a duplicate key report and an equal-width histogram — for proposal preparation records containing completed_sections and planned_sections fields.

Analyzes up to 200 proposal preparation rows to produce a duplicate key report and explicitly fill missing values in completed_sections or planned_sections fields.

Analyzes up to 200 proposal preparation records to produce a duplicate key report and numeric distribution for completed_sections or planned_sections fields.

Runs two local analytical reports on proposal preparation records: a duplicate key report and a numeric interval filter for completed_sections or planned_sections fields, across up to 200 rows.

Runs two local analytical reports on proposal preparation data: a duplicate key report and a trailing mean series for completed_sections or planned_sections fields.

Generates an equal-width histogram and fills missing values for proposal preparation records (completed_sections, planned_sections) across up to 200 rows.

Deduplicates proposal preparation records exactly and computes category frequencies for completed_sections and planned_sections fields across up to 200 rows.

Deduplicates exact rows and produces a date-by-category pivot report for proposal preparation records (completed_sections, planned_sections) — two analytical reports in one call.

Deduplicates proposal preparation records by exact row match and generates a duplicate key report for completed_sections and planned_sections data, up to 200 rows.

Deduplicates exact rows and flags IQR-based outliers in proposal preparation records (completed_sections, planned_sections) for up to 200 rows, producing 2 independent analytical reports.

Deduplicates rows and computes numeric distribution statistics for proposal preparation records with completed_sections and planned_sections fields.

Analyzes proposal preparation records by computing a grouped numeric summary and a Base-100 indexed series for completed_sections and planned_sections fields, across up to 200 rows.

Computes a grouped numeric summary and a daily sum time series for proposal preparation records (completed_sections, planned_sections) across up to 200 rows.

Generates two analytical reports (grouped numeric summary and date-by-category pivot) for proposal preparation records containing completed_sections and planned_sections fields.

Runs two local analytical reports on proposal preparation data: a grouped numeric summary by document group and a numeric interval filter on completed or planned sections, for up to 200 rows.

Generates two analytical reports for proposal preparation records: a grouped numeric summary and a numeric row ordering, based on completed_sections and planned_sections fields.

Runs IQR-based outlier detection and a date-by-category pivot analysis on up to 200 proposal preparation records (completed_sections, planned_sections) and returns two independent analytical reports.

Analyzes proposal preparation records (completed_sections, planned_sections) by computing numeric distribution statistics and adjacent-row change deltas for up to 200 rows, returning two analytical reports.

Generates a numeric distribution report and a daily sum time series for completed_sections and planned_sections fields in proposal preparation records (up to 200 rows, 2 reports per call).

Analyzes proposal preparation records by computing numeric distribution statistics and a date-by-category pivot table for completed_sections and planned_sections fields.

Generates numeric distribution statistics and fills missing values (zero/mean/median) for completed_sections and planned_sections fields across up to 200 proposal preparation rows.

Analyzes proposal preparation records by generating a numeric distribution report and a numeric interval filter report for completed_sections or planned_sections fields across up to 200 rows.

Computes a numeric distribution report and a trailing mean time series for completed_sections or planned_sections fields in proposal preparation records (up to 200 rows).

Filters proposal preparation records by a numeric interval on completed_sections or planned_sections, and computes a Base-100 index series, returning 2 analytical reports for up to 200 rows.

Applies a numeric interval filter and performs keyed dataset comparison on up to 200 proposal preparation records with completed_sections and planned_sections fields, returning 2 analytical reports.

Filters proposal preparation records by a numeric interval on a chosen field, then computes a secondary-measure weighted mean; returns two analytical reports for up to 200 rows.

Ranks rows by numeric value and generates a base-100 index series for completed_sections or planned_sections fields in proposal preparation records.

Sorts proposal preparation records by a numeric field and generates an equal-width histogram for up to 200 rows of completed/planned sections data

Sorts proposal preparation records by a numeric field and filters them to a specified numeric range, returning two analytical reports for up to 200 rows.

Computes numeric row ordering and trailing mean series for proposal preparation records (completed_sections, planned_sections) across up to 200 rows.

Runs a record quality audit and category frequency analysis on proposal preparation data (completed_sections, planned_sections) for up to 200 rows.

Runs a record quality audit and an equal-width histogram analysis on proposal preparation data (completed_sections, planned_sections) for up to 200 rows, returning two independent analytical reports.

Runs three analytical reports on proposal preparation records: a record quality audit, exact row deduplication, and category frequency counts for completed_sections and planned_sections fields.

Runs three analytical reports on proposal preparation data: a record quality audit, exact row deduplication, and a trailing mean series for completed_sections or planned_sections fields.

Audits proposal preparation records for quality issues and fills missing values in completed_sections or planned_sections fields using zero, mean, or median imputation, returning two reports for up to 200 rows.

Runs two local analytical reports — a record quality audit and a numeric distribution analysis — on up to 200 proposal preparation rows containing completed_sections and planned_sections fields.

Runs three analytical reports — record quality audit, string normalisation, and adjacent row change detection — on up to 200 proposal preparation rows containing completed_sections and planned_sections fields.

Runs three local analytical reports on proposal preparation records (completed_sections, planned_sections): record quality audit, string normalisation, and IQR-based outlier detection — up to 200 rows per call.

Runs three independent data-quality reports (record quality audit, string normalisation, and numeric interval filter) on up to 200 proposal preparation rows containing completed_sections and planned_sections fields.

Audits data quality, normalises strings, and orders rows numerically for proposal preparation records containing completed_sections and planned_sections fields.

Runs three independent analytical reports on proposal preparation records: a record quality audit, string normalisation, and two-measure correlation between completed_sections and planned_sections.

Runs two local analytical reports on proposal preparation records: string normalisation and duplicate key detection across up to 200 rows of completed/planned sections data.

Runs string normalisation and trailing mean series analysis on proposal preparation records (completed_sections, planned_sections) for up to 200 rows, returning 2 analytical reports.

Computes a trailing moving average over proposal preparation records and fills missing values explicitly, for up to 200 rows across two fields.

Computes a two-measure Pearson correlation between completed_sections and planned_sections, plus a date-by-category pivot table, for up to 200 proposal preparation records.

Computes Pearson/rank correlation between completed_sections and planned_sections, plus a keyed row-by-row comparison between two supplied proposal preparation datasets, returning two analytical reports.

Audits declared protein measurement records by comparing finite values against caller-supplied inclusive quantity/unit bounds within a specific dataset/subject/run/method/revision frame, treating null and censored values as unknown rather than zero.

Computes the difference between two ordered sequences of protein/isoform/sequence-pinned measurement records, isolating true record insertions and deletions from reordering effects.

Compares two revisions of protein/isoform/sequence-pinned measurement records to enumerate changes in owner, subject, revision, content, and scope pins by stable record ID.

Validates unit/quantity compatibility and detects conflicting unit labels within the same dataset/subject/run/method/revision/quantity frame for protein measurement records

Validates that declared counts in caller-supplied provenance manifests fall within a specified ceiling value, in matching units, without downloading sources or verifying signatures.

Counts paired missing owner, revision, and content pins in caller-supplied artifact and dataset provenance manifests without silently treating absent values as matches

Detects which records were inserted or deleted between two provenance manifest snapshots without conflating those changes with record reordering.

Enumerates changes in owner, subject, revision, content, and scope pins between two supplied provenance manifest snapshots, identified by stable record IDs.

Compares two versions of provenance manifest field declarations to detect removed properties, changed scalar types, or weakened required/non-null guarantees that would break downstream readers.

Binds caller-supplied artifact or dataset rows to a declared owner/subject/revision/content/source frame and returns a structured coverage map of provenance metadata, without verifying, downloading, or reproducing the actual source.

Fetches live and historical on-chain exchange data from the Abstract Exchange provider via Prysm's x402-monetized API gateway.

Computes a base-100 indexed time series and a secondary-measure weighted mean for publication proofing records containing checked_pages and flagged_pages data.

Computes a Base-100 indexed time series and a two-measure correlation report for publication proofing data (checked_pages, flagged_pages) across up to 200 rows.

Computes category frequency counts and a trailing moving-average series for publication proofing records (checked_pages, flagged_pages) supplied by the caller — no external lookups.

Computes daily sum time series and fills missing values (zero, mean, or median) for publication proofing records tracking checked and flagged pages, across up to 200 rows with two independent reports.

Analyzes up to 200 publication proofing records to produce a duplicate key report and fill missing values for checked_pages or flagged_pages using zero, mean, or median imputation.

Analyzes up to 200 publication proofing records to detect exact duplicate rows and identify adjacent row changes across checked_pages and flagged_pages fields.

Deduplicates publication proofing records and generates a Base-100 indexed series for checked_pages or flagged_pages across up to 200 rows.

Deduplicates exact duplicate rows and computes category-level frequency distributions for publication proofing records containing checked_pages and flagged_pages fields.

Computes a grouped numeric summary and an equal-width histogram for publication proofing records (checked_pages, flagged_pages) across up to 200 rows using local analytical engines.

Computes numeric distribution statistics and a base-100 indexed series for publication proofing data (checked_pages, flagged_pages) across up to 200 rows.

Computes numeric distribution statistics and numeric row ordering for publication proofing records (checked_pages, flagged_pages) across up to 200 rows, returning two analytical reports.

Computes a numeric distribution and a secondary-measure weighted mean for publication proofing records (checked_pages, flagged_pages) over up to 200 rows.

Applies a numeric interval filter and computes a Base-100 index series on publication proofing data (checked_pages, flagged_pages) for up to 200 rows, returning two analytical reports.