All services

Compares two submitted sample sets feature by feature and returns deterministic drift metrics. Continuous features are binned on the combined value range and receive PSI, Jensen-Shannon divergence and a Kolmogorov-Smirnov statistic; categorical features receive PSI and Jensen-Shannon divergence over exact category labels. The response names every feature, its sample counts, the metric values, the threshold used, and the reason a feature was flagged.

Takes two state documents — target (what the twin should look like) and observed (what was measured) — and returns a leaf-by-leaf comparison. Every leaf path carries one of five states: in_tolerance, out_of_tolerance, missing_in_observed, unexpected_in_observed or type_mismatch.

Research and market-monitoring agents buy this endpoint to answer a focused question with a compact live market snapshot.

Find airports near a given latitude/longitude coordinate

Get airports near a specific airport

[3-LEG] Local three-leg environment chain, one location (?loc=): leading air quality (AQI) -> current precipitation, rain and 30-day dry-wet -> realized nearest river stage and streamflow (confirmation gate). A chain, not one reading: confidence = how many legs are lit, in order. Sources: EPA/CAMS, ERA5, USGS. Descriptive only, never a forecast or advice. No data -> 422 no_charge, not billed. $0.25 per call.

Facebook ads from a specific advertiser by brand name — every live and recent ad a brand runs across Facebook, Instagram, Messenger and Threads, isolated to that ONE advertiser (filters out the stray pages a name search pulls in; echoes resolvedPage to verify). Returns the full ad creative (headline, body copy, CTA + destination link, image and video URLs), run dates, and platform mix. No Meta login. Competitor ad intelligence for AI agents.

Takes one or more district metered areas, each with a flow series and optionally a pressure series, a metered consumption figure and a service connection count, and returns three independent leak indicators per area plus a ranking across all of them. The first is minimum night flow: the lowest flow inside the night window is taken for every night in the record, the median of those nightly minima is reported, and where a connection count is given the legitimate night consumption is subtracted so…

Ereignisfeed für Dunkelflaute, Sturm, negative Preise und Rekorde.

Energy-trading and flexible-load agents buy this endpoint to compare the next 24 hours of EPEX prices and volatility.

Flexible-load agents buy this endpoint to schedule a contiguous operating window at the lowest average electricity price.

Takes an ordered list of positions with timestamps, optionally carrying the battery state of charge at each fix, and returns what a fleet operator asks of a trace: total distance over ground, how much of the elapsed time the vehicle was moving and how much it stood, each individual stop with its position and duration, the mean and maximum speed, and — where a battery capacity is supplied — the energy drawn and the consumption per kilometre.

Fits a constant-elasticity demand curve to the price and sales series you supply and returns the elasticity with its standard error, t statistic, two-sided p value and a Student t confidence interval, not a normal one, because at eight observations the difference in interval width is double-digit percent and a too narrow interval is exactly the kind of number a machine reuses unchecked.

Takes a list of candidate intents, each described by attributes, plus whatever is already known, and returns the one attribute worth asking about next. Candidates contradicted by the known values are ruled out first and listed with the attribute that excluded them; a candidate that simply does not declare a known attribute is kept, because silence is not a contradiction.

Takes a message history and a character budget and returns a shortened history together with a full account of what it cost. Four stages run in fixed order: pin protects messages that must survive (roles listed in keep_roles, the last keep_last_n messages, and anything you marked pin), dedupe removes later near-duplicates of an earlier message of the same role using word-overlap similarity, trim shortens unprotected survivors to max_message_chars at the last sentence boundary, and drop removes…

List recent JPYC Transfer events on one supported chain from roughly the last hour of blocks, optionally filtered by address — tx hash, from, to, amount, block. Without a cursor: newest first. With a cursor: only events after that position, oldest first, so each observed event is returned once within the scanned window. Use to confirm a recent JPYC payment or monitor wallet activity. Not for history beyond the window, sanctions screening, identity, or finality proof.

Query onchain DEX token price & market data based on the provided token contract addresses on a network.

Walks the events you pass through the ordered steps you name and reports per step: how many arrived, conversion from the previous step and from the first, drop-off in users and as a rate, and time to the next step as median, p90 and mean. A step counts only if it happened after the one before and inside the window. Where events carry a segment, each is tested against the largest with the same two-proportion test the A/B endpoint uses. Nothing is stored.

Turns SenML (JSON or CBOR) and LwM2M TLV into one record shape: resolved name, unit, absolute timestamp, typed value. Optionally puts every series on a shared time axis with a chosen interval and fill rule (last value, bucket mean, linear). Works only on the payload you pass: no device is contacted, nothing is stored, the same request returns the same bytes. TLV carries no type on the wire; unmapped resource ids come back as opaque base64 with a warning instead of a guess.

Builds the cohort table from the events you pass: users grouped by the period of their first event, counted again in each following period. Returns the table per cohort, a retention curve across all cohorts, and, where events carry a value, cumulative value and value per user. The curve counts only cohorts old enough for a given period, so a young cohort is not scored as churn. Day, week and month are real calendar periods. Nothing is fetched, nothing is stored.

Takes a machine definition, the current state, one event and an optional context, and computes a single step. Transitions may carry a guard, which is a declared comparison — field, operator, value — and never an expression: the field is a dotted path into context or payload, so a machine handed in by a stranger cannot execute anything on this server.

Calculates item co-occurrence and item similarity from the submitted events only. No user profile is stored.

Reads one conversion experiment two ways at once. Frequentist: two-proportion test, p value, confidence interval on the absolute lift, and the sample size the requested minimum detectable effect needs at the requested power. Bayesian: Beta-Binomial posteriors, probability the variant is better, expected loss either way, credible intervals - closed form, not sampled, so the numbers repeat exactly. looks_taken carries the adjusted threshold and the rate you actually ran at.

X402 Resource Match: X402 Resource Match compares challenge resource URL with the intended paid resource from bounded caller-supplied values without an external provider. Call X402 Resource Match before paying for or accepting an x402 or agent delivery assembled from caller-supplied challenge and result metadata. Returns contract-specific checks, normalized evidence, mismatch reasons, and an explicit acceptance or rejection status for X402 Resource Match as versioned deterministic JSON. Price:…