Every agent-payable API.

Gross profit - Record ranking: Rank up to 200 supplied records by calculated value, preserving row indices; higher is not necessarily better. Metric: Gross profit. Supplied figures only; returns JSON with formula, units and limitations. One of 50 calculators with 20 analysis modes. No live feed, financial advice or trade execution. 0.01 USDC includes one bounded record ranking report.

Report the ground resolution (metres per pixel, for 256-px tiles) and the OGC map scale denominator at a zoom level, optionally adjusted for latitude (default equator).

OpenAI-compatible chat completions GROUNDED in a live web search on every call: the gateway runs an Exa search (up to 5 results) for the prompt, hands the results to the model, and returns the answer with url_citation annotations - $0.03 per call in USDC, no API key, no signup. Model is chosen server-side like the auto tier (omit "model" or send "auto"; explicit ranked models accepted); the response adds agent402_router {category, quality, served}.

Format a card number into the conventional display groups for its network (4-6-5 for Amex, 4-6-4 for 14-digit Diners, otherwise groups of four).

Grove consented-data pilot delivery. Requires an owner-enabled offer query parameter, the agreed license hash, and payment from its named buyer wallet. 25 data templates and 12 use types. One purchase per offer, 0.01 USDC to Grove Shared Fund. No personal records are exposed in discovery. Obtain an offer from its contributor before paying.

GTIN / EAN / UPC barcode check-digit validation (GS1 mod-10) for GTIN-8/12/13/14 with type identification.

Extract a public web page into timestamped, normalized evidence.

Creates or inspects H3 cells and optionally returns neighbors or changes resolution. Missing facts are returned as null and named in unavailable_fields where applicable. Derived values disclose their method; source-backed values disclose source, time window and age.

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.

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.

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.

Runs three detectors over one series and reports what each saw: a robust z score on median and median absolute deviation, a robust seasonal-trend decomposition with a z score on the residual, and an isolation forest over value and step-to-previous. Every score is returned per point, so callers set their own threshold instead of trusting a verdict. The forest draws from a fixed seed: same request, same result. Under three full cycles the seasonal figure is marked as not robust.

Aktuelle CO2-Intensität des deutschen Strommix.

Cheapest end-to-end x402 purchase test. Always returns a 402 challenge without payment and the current full NegPrice decision after a valid USDC payment.

Trading and dispatch agents buy this endpoint to obtain the current German EPEX spot price for immediate decisions.

Battery, EV, and trading agents buy this endpoint to receive calibrated 2h/4h/6h negative-price probabilities, reviewed threshold profiles and an executable response policy.

Combined share of permanent water, herbaceous wetland and mangrove classes. The service evaluates a 60 metre local analysis grid around one coordinate and returns standardized units across all 30 covered countries.

Checks one address on Ethereum, Base, Arbitrum, Optimism, Polygon or BNB Smart Chain against the open verification registry and against the chain itself. For a proxy the answer is deliberately split in two: whether the proxy is verified says almost nothing, whether the implementation it currently points to is verified is the real question, and both are reported separately with the implementation address read from the EIP-1967 storage slot, the EIP-1167 clone pattern or an implementation() call —

Reads the public perpetual data of Binance USDⓈ-M Futures, Bybit linear and OKX swaps for one BASE/QUOTE pair. Settlement intervals differ per exchange and per contract, so annualized_pct is only computed after normalising to the interval that exchange actually uses, and the formula is part of the answer. current_rate and predicted_rate are strictly separate: where an exchange publishes no forecast, predicted_rate stays null and the reason is named instead of reusing the current rate. next_settl

Returns the raw public trade tape of Binance, Coinbase, Kraken, OKX or Bybit for one BASE/QUOTE pair, oldest first, exactly as the exchange reports it — no smoothing, no merging of prints. side is the side of the aggressor: reported directly by Kraken, OKX and Bybit, derived from the maker flag at Binance, and inverted from the maker side at Coinbase; side_source names which case applies. Where an exchange does not publish it, side is "unbekannt" and side_inferred stays false rather than guessin

Returns the visible order book of Binance, Coinbase, Kraken, OKX or Bybit for one BASE/QUOTE pair. Levels come as [price, size] with bids descending and asks ascending. depth[] states how much turnover sits within 0.5, 1 and 2 percent of the mid per side, in the quote currency, and additionally as depth_usd where the quote is a US dollar unit. group_bps folds levels into price bands whose price is the size weighted average of the band, for agents that do not want a thousand rows. The exchange up

Answers the timing question an agent has before it schedules anything against a regulated exchange: is the main session running, and if not, when does it start. Covers XETR, XFRA, XNYS, XNAS, XLON, XPAR, XAMS, XSWX, XTKS and XHKG with pre trading, main session, post trading, half days and the holidays of 2026 and 2027. Time zones are resolved through IANA names and Intl at the queried instant, never through a fixed offset to UTC, because summer time starts and ends on different dates in the Euro

Asks every known facilitator what it supports and measures how long that answer takes. Reachability, supported chains, supported assets, supported schemes and the current response time are measured live on every call. Limits, and they are the important part: error rate over 24 hours, uptime over 7 days and latency percentiles CANNOT be delivered. They require a series of own measurements every few minutes, and this service keeps no such series - a single call cannot produce a p95. Those fields s

Checks vouchers of a payment channel by computation only - nothing is ever claimed, redeemed or refunded. Signatures are verified cryptographically against the address each voucher claims to be signed by, in EIP-712 form when the typed data is supplied, otherwise as a plain message signature. Double spending is the expensive case and is checked in full: every voucher id occurring more than once is reported with all its occurrences and whether the duplicates agree on the amount. Sums are added up

Compute a keyed HMAC of some data with a secret key — the standard way to sign and verify webhooks (Stripe, GitHub, etc.) and API requests. Data and key may each be utf8/hex/base64/base64url; the tag is returned as hex/base64/base64url.

HMAC of a message under a key — Genesis402 / UnyKorn Operator Network

Generate an RFC 4226 counter-based HMAC one-time password from a base32 secret and counter. Configurable digits (6–10, default 6) and algorithm (SHA1/SHA256/SHA512, default SHA1). An invalid base32 secret is a clean 400.

Map a short product description to likely HS chapter/heading hints (electronics, apparel, coffee, pharma, …). Keyword table only — not a binding customs classification; confirm before filing.

Call when an agent needs text decoded with html-named. No network and no LLM. Exact $0.008 USDC. Prefer unpaid POST /v1/sandbox/html-named-decode first.

Call when an agent needs text encoded with html-named. No network and no LLM. Exact $0.007 USDC. Prefer unpaid POST /v1/sandbox/html-named-encode first.

Call when an agent needs text decoded with html-numeric. No network and no LLM. Exact $0.01 USDC. Prefer unpaid POST /v1/sandbox/html-numeric-decode first.

Call when an agent needs text encoded with html-numeric. No network and no LLM. Exact $0.009 USDC. Prefer unpaid POST /v1/sandbox/html-numeric-encode first.

Call when an agent needs HTML sanitized to an allowlist of tags and safe URLs. No network. Exact $0.003 USDC. Prefer unpaid POST /v1/sandbox/html-sanitize first.

Convert an HTML document to clean Markdown for LLM ingestion: strips scripts, styles and navigation chrome, preserves headings, lists, tables and links. Deterministic for identical input.

Call when an agent already has HTML and needs clean markdown. No network. Optional url is a base for relative links only. Exact $0.002 USDC. Prefer unpaid POST /v1/sandbox/html-to-md first.

Convert a supplied HTML string into deterministic, LLM-ready Markdown. Preserves headings, links, lists, emphasis, and code without fetching a URL or invoking an AI model. Input is bounded to 2,000,000 characters.

Call when an agent needs HTML converted to plain text (scripts and styles removed). No network. Exact $0.001 USDC. Prefer unpaid POST /v1/sandbox/html-to-text first.

Decode HTML entities (named, decimal and hex numeric references) back into plain text.

Encode text to HTML entities so it can be safely embedded in markup. By default uses hex numeric references; set useNamedReferences for named entities (&, ©) or encodeEverything to encode every character.

SSRF-safe public GET with status, headers, and size-capped JSON/text body for agent tool use

Public URL status, latency, and response header subset for uptime and health agents

Public URL status, latency, and response header subset for uptime and health agents

Call when an agent needs response headers from an SSRF-safe HEAD or GET of a public URL. Exact $0.003 USDC. Prefer unpaid POST /v1/sandbox/http-headers first.

HTTPX Declared dependencies: Live PyPI declared dependencies for HTTPX. Package name can be overridden.

HTTPX Latest release summary: Live PyPI latest release summary for HTTPX. Package name can be overridden.

HTTPX Recent release history: Live PyPI recent release history for HTTPX. Package name can be overridden.

Hacker News full-archive search: stories, comments, Show HN and Ask HN with points, comment counts and URLs

Search Hacker News stories and comments by query, sorted by relevance or date (Algolia HN).

Hacker News search: stories or comments, by relevance or recency — Genesis402 / UnyKorn Operator Network