All Services

Runs two local analytical reports on label review data: a numeric interval filter and a daily sum series, covering up to 200 rows of reviewed_records or changed_labels.

Runs two analytical reports on label review records — a numeric interval filter and an equal-width histogram — for up to 200 rows of reviewed_records or changed_labels data.

Applies a numeric interval filter and IQR-based outlier detection to label review records (reviewed_records, changed_labels), returning two analytical reports for up to 200 rows.

Analyzes label review records to produce two reports: numeric row ordering and adjacent row changes for reviewed_records or changed_labels fields, across up to 200 rows.

Sorts label review records by numeric value and flags statistical outliers using the interquartile range (IQR) method, producing two independent analytical reports for up to 200 rows.

Analyzes label review records by sorting rows numerically and comparing two keyed datasets to surface differences in reviewed_records and changed_labels metrics.

Sorts label review records by a numeric field and filters rows within a specified numeric interval, returning two analytical reports for up to 200 rows.

Computes numeric row ordering and secondary-measure weighted mean statistics for supplied label review records (reviewed_records and changed_labels) across up to 200 rows.

Runs two analytical reports on label review data: a record quality audit and an adjacent-row change analysis, for up to 200 rows of reviewed_records and changed_labels data.

Runs two local analytical reports on label review data: a record quality audit and category frequency analysis for reviewed_records and changed_labels, covering up to 200 rows.

Runs two local analytical reports on label review data: a record quality audit and an equal-width histogram, for up to 200 rows of reviewed_records or changed_labels values.

Runs three local analytical reports on label review data: record quality audit, exact row deduplication, and category frequency breakdown — all in one call for up to 200 rows.

Runs three local analytical reports on label review data: a record quality audit, exact row deduplication, and an equal-width histogram — all computed offline against up to 200 supplied rows.

Runs three bundled analytical reports on label review data: a record quality audit, exact row deduplication, and numeric distribution analysis for reviewed_records or changed_labels.

Runs a two-report analytical audit on label review data: a record quality audit and IQR-based outlier flagging for up to 200 rows of reviewed_records and changed_labels data.

Performs a dual-report analysis on label review data: a record quality audit and a keyed dataset comparison between two supplied datasets of up to 200 rows each.

Runs three independent analytical reports (record quality audit, string normalisation, and adjacent row changes) on up to 200 label review records containing reviewed_records and changed_labels fields.

Runs three analytical reports on label review data: record quality audit, string normalisation, and category frequency analysis for up to 200 rows of reviewed records and changed labels.

Runs three local analytical reports (record quality audit, string normalisation, and duplicate key detection) on up to 200 rows of label review data containing reviewed_records and changed_labels fields.

Audits label review records for data quality issues, normalises string fields, and fills missing values using zero/mean/median imputation for up to 200 rows across reviewed_records and changed_labels fields.

Runs three analytical reports (record quality audit, string normalisation, and numeric distribution) on up to 200 rows of label review data containing reviewed_records and changed_labels fields.

Runs three local analytical reports on label review data: record quality audit, string normalisation, and numeric interval filtering for reviewed_records or changed_labels fields across up to 200 rows.

Runs two analytical reports (record quality audit and trailing mean series) on label review data containing reviewed_records and changed_labels fields, for up to 200 rows.

Performs string normalisation and category frequency analysis on label review records (reviewed_records, changed_labels) for up to 200 rows, returning two independent analytical reports.

Runs two local analytical reports on label review data: string normalisation and exact row deduplication, for up to 200 rows of reviewed_records and changed_labels data.

Runs string normalisation and keyed dataset comparison on two sets of label review records (reviewed_records, changed_labels), returning 2 independent analytical reports for up to 200 rows each.

Analyzes up to 200 label-review records, producing a string normalization report and a numeric distribution report for the reviewed_records or changed_labels field.

Computes two analytical reports (trailing mean series and adjacent row changes) on label review data (reviewed_records or changed_labels) for up to 200 rows.

Computes trailing moving-average series and base-100 index series for label review records (reviewed_records and changed_labels) across up to 200 rows.

Computes a trailing moving-average time series and a date-by-category pivot table for label review data (reviewed_records and changed_labels) supplied by the caller.

Computes a trailing mean time series and an equal-width histogram for label review data (reviewed_records or changed_labels) across up to 200 rows, returning two independent analytical reports.

Computes trailing mean time-series and IQR-based outlier flags for label review records (reviewed_records and changed_labels) across up to 200 rows, producing two analytical reports.

Computes a trailing mean time series and a two-measure Pearson-style correlation for label review data (reviewed_records vs changed_labels) over up to 200 rows.

Computes Pearson/Spearman correlation between reviewed_records and changed_labels, and performs a keyed row-by-row comparison between two label review datasets, returning two independent analytical reports.

Calculates Pearson correlation between labour force participation rates and user-supplied benchmark figures, returning a JSON report with formula, units, and limitations.

Calculates the local finite-change elasticity of labour force participation rate with respect to a small percentage change in a specified input variable.

Calculates a weighted mean of labour force participation rates across up to 200 supplied data rows using user-provided nonnegative weights.

Calculates a base-100 rebased index of landed unit cost across multiple rows of retail cost data, enabling scenario comparison relative to a chosen baseline row.

Computes statistical benchmark correlation for landed unit cost figures across multiple retail data rows against supplied benchmark values.

Computes a statistical distribution summary of landed unit costs across multiple rows of retail cost data (goods, freight, duty, other costs, units).

Calculates IQR-based outlier flags for landed unit cost across a batch of retail cost rows supplied by the caller.

Calculates and analyzes ordered trends in landed unit cost across a sequence of cost data rows, including goods, freight, duty, and other costs per unit.

Ranks multiple product records by their calculated landed unit cost (goods cost + freight + duty + other costs divided by units), returning a sorted leaderboard of cost efficiency.

Calculates the gap between a target landed unit cost and the actual landed unit cost derived from supplied goods cost, freight, duty, other costs, and unit count.

Classifies landed unit cost values against configurable lower and upper thresholds for retail scenario analysis.

Purchases $5 USDC worth of LD (LAND) tokens at contract address 0x5fC5484b8dbaBf274b6CDd3Ec3E280B1d8cA2B07 and returns transaction details

Computes confidence calibration bins for a language routing classifier given up to 200 labelled text records with actual labels, predicted labels, and class probabilities.

Computes a fixed-holdout learning curve for a naive-Bayes language routing classifier (english/french/spanish/other) using up to 200 supplied labelled text records.

Runs a grouped three-fold cross-validation using a naive-Bayes baseline model on up to 200 labelled text records to evaluate language routing accuracy across English, French, Spanish, and Other classes.

Computes the multiclass Brier score for a language routing classifier given up to 200 labelled text records with predicted probability distributions.

Computes multiclass log loss metric for language routing model predictions across English, French, Spanish, and other language classes from up to 200 labeled text records.

Computes per-label precision and recall metrics for a language routing classifier across English, French, Spanish, and Other categories using up to 200 labelled text records.

Evaluates the accuracy of a language routing classifier across supplied labelled text records, broken down by group membership.

Computes top-k label coverage metrics for a language routing classifier given up to 200 labelled text records with predicted probabilities.

Trains a naive-Bayes text classifier on up to 200 labelled records (english/french/spanish/other) and classifies new query texts against that local model.

Searches academic literature on large language model reliability using Crossref metadata, returning DOI references and source attribution.

Returns a breakdown of academic publications on large language model reliability, with DOI references and source attribution via Crossref publisher metadata.

Retrieves recent academic publications on large language model reliability using Crossref publisher metadata, returning DOI references and source attribution.

Retrieves scholarly citations and funding attribution for academic research on large language model reliability using Crossref metadata

Returns structured citation records (DOI references and source attribution) for academic literature on large language model reliability, using Crossref publisher metadata.

Calculates period-over-period changes in the largest reported token holder's share of circulating supply across a series of observations.

Calculates the largest reported token holder share (as a percentage of circulating supply) for a batch of up to 200 token observations in a single call.

Computes the Pearson or equivalent benchmark correlation between a series of largest-token-holder share observations and a supplied benchmark series.

Computes a distribution histogram of the largest reported token holder share percentages across a supplied set of token observations.

Computes a statistical distribution summary of the largest reported token holder share (holder tokens / circulating tokens) across multiple supplied observations.

Computes the local finite-change elasticity of the largest reported token holder's share with respect to small perturbations in holder tokens or circulating tokens.

Computes a one-input sensitivity table showing how the largest reported token holder share metric changes as either holder_tokens or circulating_tokens varies across a supplied range of values.

Analyzes ordered trend patterns in the largest reported token holder's share (holder tokens as a fraction of circulating tokens) across a sequence of observations.

Ranks up to 200 supplied token holder share observations by calculated holder percentage, returning a deterministic record ranking report.

Computes a rolling mean of the largest reported token holder's share (holder_tokens / circulating_tokens) over a sliding window across a series of supplied observations.

Calculates how a change in holder_tokens or circulating_tokens sequentially shifts the largest-holder share percentage, attributing each input's contribution to the overall delta.

Calculates the percentage share of the largest token holder given their token count and total circulating supply.

Calculates the gap between the current largest token holder share and a specified target share percentage.

Classifies whether the largest reported token holder's share falls within a user-defined threshold range, given token holder and circulating supply data.

Computes a two-dimensional scenario grid of the largest token holder's share percentage across varying holder_tokens and circulating_tokens values.

Computes a weighted mean of the largest reported token holder share (holder_tokens / circulating_tokens) across up to 200 supplied observations.

Issues an access token upon receipt of an exact 1000 USDC (v2) payment on Base (eip155:8453) within a 300-second window

Compares two Last-Modified HTTP timestamps (as Unix epochs) and computes their delta against caller-supplied bounds, returning a pass/advisory status with normalized evidence.

Executes one play of the sequence arcade game at The Latent Lounge, returning the result of a single game round

Submit an answer to the daily oracle question at The Latent Lounge, permanently archiving the response on-chain

Returns an educational briefing on 2026 model-year cars covering powertrains, range, pricing, and new features.

Purchases and delivers the Terminal design pack for a single project, covering dark-first UI design guidelines for developer tools, API consoles, and docs sites.

Computes confidence calibration bins for a multi-class lead stage classifier (enquiry, qualified, negotiation, closed) from up to 200 labelled records with predicted probabilities.

Computes a confusion matrix for a lead stage classifier across four stages (enquiry, qualified, negotiation, closed) from up to 200 labelled text records.

Generates a fixed-holdout learning curve for a naive-Bayes text classifier across four lead pipeline stages (enquiry, qualified, negotiation, closed) using up to 200 labelled records.

Audits classifier labels and predictions for lead stage classification (enquiry, qualified, negotiation, closed) using local naive-Bayes evaluation tools on up to 200 supplied records.

Computes macro, micro, and weighted classification evaluation scores for a lead-stage prediction dataset with up to 200 labelled text records.

Computes the multiclass Brier score to evaluate probabilistic classifier performance on lead stage predictions across four sales pipeline stages.

Computes multiclass log loss for lead stage classification models across up to 200 labelled text records with enquiry/qualified/negotiation/closed categories.

Evaluates a user-supplied set of labelled lead-stage text records by group, returning per-group classification metrics and naive-Bayes baselines for the enquiry/qualified/negotiation/closed taxonomy.

Generates a taxonomy prompt template for classifying lead stage text records into enquiry, qualified, negotiation, or closed categories using up to 200 labelled examples.

Computes top-k label coverage metrics for a naive-Bayes classifier over the four lead pipeline stages (enquiry, qualified, negotiation, closed)

Returns a breakdown of a sample of academic publications on learning analytics, with DOI references and source attribution from Crossref publisher metadata.

Returns reported research funders for learning analytics publications using Crossref publisher metadata, with DOI references and source attribution.

Returns a structured decision procedure for agents whose stored API key has died and must decide whether to verify, reroute, migrate to x402 pay-per-call, or fail over gracefully

Compares and evaluates trust verdicts for agents or resources in the x402 agentic economy, settled on Algorand

Returns a heatmap of whale wallet activity and large transaction concentration across blockchain addresses or tokens

Fetches real-time GLP (GMX Liquidity Pool) statistics and analytics for the GMX perpetuals protocol on Arbitrum.

Retrieves the on-chain balance for a given wallet address via the LensCrypt crypto intelligence API

Retrieves NFT sales data for a specific contract address on a given blockchain chain