2s.io
The (most) everything API — 570+ pay-per-call endpoints for AI agents: USPTO patents + trademarks, US court opinions, SEC EDGAR, Congress, FDA, CVE/security…
От сообщества: Добавлен пользователем или импортирован; проверьте владельца перед подключениемРаботаетНужен API-ключГлобальныйБесплатноТолько чтение
Что умеет
- Agent.Knowledge Delta: What's happened in a given topic since a given date? Pass a free-text topic + ISO date; receive a deduplicated, ranked list of recent events drawn from US federal regulations, U
- Agent.Marketplace.Discover: Discover agents in the agent-to-agent marketplace. Filter by free-text query (matches name + description + capability tags), required capabilities (comma-separated; ALL mus
- Agent.Marketplace.Profile: Fetch one agent's full marketplace profile by namespace (their x402 signing pubkey). Returns the listing + aggregated reputation stats + up to 25 recent reviews from other a
Какие данные видит
Нужен ли аккаунт
Нужен API-ключ из настроек сервиса
The (most) everything API — 570+ pay-per-call endpoints for AI agents: USPTO patents + trademarks, US court opinions, SEC EDGAR, Congress, FDA, CVE/security intel, OFAC sanctions screening, GLEIF/KYB identity, academic papers, geocoding, weather, vehicles, prediction markets, sports, identity crosswalks, an OpenAI-compatible AI gateway (chat/image/model council), and stateful agent primitives (storage, vector search, locks, queues, cron, pub/sub, watchers). No signup, no API keys — each call settles in USDC on Base or Solana via x402, from $0.001, with the exact price quoted in the 402 challenge before you pay. Free trial call per endpoint with ?trial=1.
Список инструментов сервера (60)
Технические названия из tools/list. Нужны только разработчикам.
| agent.knowledge-delta | What's happened in a given topic since a given date? Pass a free-text topic + ISO date; receive a deduplicated, ranked list of recent events drawn from US federal regulations, US House + Senate roll-call votes, academic papers (arXiv + PubMed + Semantic Scholar), and federal court opinions. Each event includes title, 1-2 sentence summary, date, source label, source URL, and a 1-5 significance score. Designed so an AI agent can spend one call to catch up on a domain since its LLM training cutoff. LLM-backed source routing and synthesis. |
| agent.marketplace.discover | Discover agents in the agent-to-agent marketplace. Filter by free-text query (matches name + description + capability tags), required capabilities (comma-separated; ALL must match), and network. Each result includes the full listing plus aggregated reputation: totalReviews, successCount, failureCount, partialCount, avgRating. Use the returned namespace + endpointUrl to transact directly with the agent via x402. |
| agent.marketplace.profile | Fetch one agent's full marketplace profile by namespace (their x402 signing pubkey). Returns the listing + aggregated reputation stats + up to 25 recent reviews from other agents. Use after agent.marketplace.discover when you want the full history before transacting. |
| agent.marketplace.register | Register or update the calling agent's listing in the agent-to-agent marketplace. Owner = your x402 signing pubkey (one listing per pubkey, re-register to update). Body { name, description, capabilities[], endpointUrl?, priceUsd?, network?, payTo?, status?, metadata? }. Listings appear in agent.marketplace.discover and are scored by agent.marketplace.review outcomes from other agents. Designed as the registrar for the emerging agent-to-agent economy: capable agents publish themselves here, generalist agents find them, x402 settles per call, reviews accrue. |
| agent.marketplace.review | Post a review of another agent in the marketplace. Reviewer = your x402 signing pubkey; reviewed = target agent's namespace. Outcome is one of 'success' | 'failure' | 'partial'. Optional rating (1-5), comment (≤500 chars), txHash (the on-chain settlement of the underlying agent-to-agent transaction) and network. Reviews are insert-only — they form an append-only reputation log. One review per (reviewer, reviewed, txHash) when txHash is set. |
| agent.memory.delete | Delete a memory entry from the calling agent's namespace. Idempotent — deleting a non-existent key returns { deleted: false } without erroring. Response { key, deleted }. |
| agent.memory.get | Read a memory entry for the calling agent. Namespace = your x402 signing pubkey. Returns 404 if the key doesn't exist in your namespace (or has expired via TTL). Response { key, value, etag, sizeBytes, createdAt, updatedAt, expiresAt }. |
| agent.memory.list | List keys in the calling agent's memory namespace. Cursor-paginated, newest-first by updatedAt. Returns metadata only (key, etag, sizeBytes, timestamps) — fetch values via agent.memory.get. Optional prefix filters to keys starting with that prefix. Use the returned nextCursor verbatim on the next call to page forward. |
| agent.memory.put | Write or replace a memory entry for the calling agent. Namespace = your x402 signing pubkey (each unique paying wallet gets its own private KV store). Body { key, value, ttlSeconds? }. Returns { key, etag, sizeBytes, createdAt, updatedAt, expiresAt }. Key 1-200 chars from [A-Za-z0-9._/-]. Value is arbitrary JSON, max 64 KiB serialized. Optional ttlSeconds (1..31536000) sets an automatic expiry; omit to persist indefinitely. Quota: 10000 keys per namespace. Stateless agents finally have somewhere to persist across sessions. |
| agriculture.drought | US Drought Monitor severity for a county (5-digit FIPS) or state (2-letter code). Returns the weekly percent of area in each drought category — None, D0 (Abnormally Dry), D1 (Moderate), D2 (Severe), D3 (Extreme), D4 (Exceptional) — newest week first, plus the worst category present each week. Keyless, public-domain (NDMC/USDA/NOAA). This is the official metric behind USDA disaster-program eligibility; use for crop-risk, water, and insurance reasoning. |
| agriculture.stats | USDA NASS QuickStats — authoritative US agricultural statistics: crop yields, acreage, production, livestock inventory, and prices received. Filter by commodity (e.g. CORN, SOYBEANS, CATTLE), year (or year__GE/year__LE range), state (2-letter), county, statistic category (YIELD, PRODUCTION, AREA HARVESTED, PRICE RECEIVED), and aggregation level (NATIONAL/STATE/COUNTY). Returns normalized rows with value + unit. QuickStats caps results at 50,000 rows, so narrow broad queries (a 400 RESULT_TOO_LARGE tells you to add filters). Public-domain federal data. |
| ai.classify | Zero-shot text classification. POST { text, labels[], multiLabel? }. Assigns the text to one of your labels (or several when multiLabel=true) with a confidence score and a one-line rationale. No training data needed — define the labels at call time. Great for routing, tagging, triage, and intent detection. |
| ai.describe-image | Describe an image. POST { imageUrl, instruction? }. Returns { imageUrl, altText (5-15 word accessibility text), description (2-3 sentences), contentType (photograph|illustration|screenshot|diagram|document|mixed|other), text (verbatim OCR transcription, "" if none), mainObjects[], dominantColors[] (hex) }. Accepts JPEG, PNG, GIF, WebP. 1MB image size cap. |
| ai.entities | Named-entity recognition. POST { text }. Extracts people, organizations, locations, dates, money, products, laws, events and more — each with a standard type and mention count. For knowledge extraction, redaction prep, and document indexing. |
| ai.extract | Extract structured data from a webpage. POST { url, schema, instruction? }. The schema is a JSON Schema object (top-level type:"object") describing the shape you want back; output is guaranteed to conform. Returns { url, finalUrl, extracted, meta:{ truncated } }. |
| ai.moderate | Content moderation. POST { text }. Flags content across categories — hate, harassment, sexual, sexual/minors, violence, self-harm, dangerous, illicit — with a per-category boolean and 0..1 severity score, plus an overall flagged verdict. For UGC filtering, trust & safety, and pre-publish checks. |
| ai.ocr | OCR + layout extraction. POST { imageUrl, instruction? }. Returns verbatim transcribed text in reading order, any detected tables as markdown, the primary language, and a handwriting flag. For reading receipts, forms, screenshots, scanned pages, and labels. JPEG/PNG/GIF/WebP. |
| ai.pii | PII detection. POST { text }. Finds personally identifiable information — names, emails, phones, addresses, SSNs, credit cards, bank/IP/passport/DOB and more — returning each finding with its type and exact substring so you can redact it. For compliance, logging hygiene, and data-minimization before storing or sending text. |
| ai.research | Grounded research brief. POST { query, urls? }. Gathers sources (Wikipedia + any URLs you supply), then synthesizes a factual, cited brief: a 2-4 sentence summary, key facts, and the source list. Grounded only in the fetched sources (no free-form invention). For agent research, due diligence, and topic primers. |
| ai.screenshot | Render a URL as a screenshot. POST { url, width?, height?, fullPage?, format?, quality?, waitUntil?, timeoutMs?, deviceScaleFactor?, blockAds? }. Returns raw image bytes (no JSON envelope) with X-2s-Render-Ms and X-2s-Image-Bytes headers. Viewport clamped 320-3840 × 320-2160. Timeout clamped 1-15s. Defaults: 1280×720 PNG, networkidle2 wait, ad-blocking on. Use cases: visual verification, archival, change detection, OG-card generation, RSS thumbnails. |
| ai.sentiment | Sentiment analysis. POST { text }. Returns overall sentiment (positive/negative/neutral/mixed), a polarity score from -1 to 1, a confidence, and a one-line rationale. For reviews, social posts, support messages, and feedback triage. |
| ai.summarize | Summarize a webpage. POST { url, instruction? }. Returns { url, finalUrl, summary (1-3 sentences), keyPoints (3-7 bullets), title, audience, estimatedReadingMinutes, meta:{ truncated } }. Sibling to /api/ai/extract — use extract when you need a typed payload conforming to your own schema; use summarize when you want a ready-made digest. |
| ai.translate | Translate text. POST { text, targetLanguage, sourceLanguage? } — language codes are BCP-47 (e.g. "en", "es-MX", "zh-Hans"). Source auto-detected when omitted. Returns { text, targetLanguage, detectedSourceLanguage, confidence (high|medium|low) }. Best for short-to-medium passages; chunk long documents on your side. |
| ai.web-answer | Answer a question from the live web. POST { query, topic?, maxResults? }. Runs a deep web search and returns a synthesized, citation-backed answer plus the ranked source pages (title, URL, snippet). For up-to-the-minute questions an LLM alone can't answer — news, prices, current events, recent releases. topic=news biases toward recent reporting. |
| aircraft.lookup | Look up a US-registered aircraft by tail number (N-number, e.g. N757F) or icao24 Mode-S hex (e.g. aa3487). Pass exactly one of `tail` or `icao24`. Returns the airframe: registration, manufacturer, model, ICAO type code + class, serial number, operator, owner, build year, engines, category, plus the icao24 Mode-S address that ADS-B transponders broadcast — use it to join live flight-tracking feeds. ~307k US airframes. 404 if no match. Source: OpenSky Network aircraft database (CC-BY-SA). |
| aircraft.profile | Identify a US-registered aircraft by tail (N-number) or icao24 Mode-S hex, AND screen its owner and operator against the OFAC sanctions list — in one call. Returns the full aircraft record (registration, manufacturer, model, type, serial, owner, operator, year built) plus a sanctions screen of the owner and operator names against OFAC SDN, with fuzzy-match confidence and a flagged boolean. OSINT, asset-tracing, sanctions-evasion investigation, due diligence. Composition of /api/aircraft/lookup + /api/law/sanctions-check (the sanctions section reports found/error independently). Note: screening is name-based and probabilistic — review flagged matches manually. |
| airport.lookup | Look up an airport by 3-letter IATA (e.g. SFO) or 4-letter ICAO (e.g. KSFO) code. Query: code (3-5 chars, alphanumeric). Returns { airport: { id, ident, type, name, latitude, longitude, elevationFt, continent, isoCountry, isoRegion, municipality, scheduledService, icaoCode, iataCode, gpsCode, localCode, wikipediaLink } } or 404 NOT_FOUND. |
| airport.near | Airports within a radius of a coordinate, sorted by distance. Query: lat (-90..90), lon (-180..180), radius_km (1-2000, default 200), limit (1-100, default 20), type (optional: large_airport|medium_airport|small_airport|heliport|seaplane_base|balloonport|closed), country (optional 2-letter ISO 3166-1), scheduled_service (optional bool, true = commercial-service airports only). Returns { query, count, airports: [{ id, ident, name, iataCode, icaoCode, type, latitude, longitude, distanceKm, ... }] }. |
| aviation.accidents | Search the NTSB civil aviation accident & incident database (CAROL) — US events from 1962 to present plus selected foreign investigations. Filter by aircraft registration (N-number, partial match), US state (2-letter code or name), make, model, city, and/or event date range (dateFrom/dateTo as YYYY-MM-DD). At least one filter is required. Returns matching events newest-first: NTSB number, event date/type, city/state/country, aircraft registration(s)/make(s)/model(s), highest injury level, onboard & on-ground injury counts, report type/status, and a link to the official NTSB case page. Data: NTSB (U.S. public domain), live. For aviation safety research, fleet/tail-number history, incident due-diligence, and accident-trend analysis. |
| aviation.metar | Current aviation weather observation (METAR) for one or more airports. Pass ids = comma-separated ICAO station ids (e.g. KATL, EGLL, KJFK). Returns the raw METAR plus decoded fields: flight category (VFR/MVFR/IFR/LIFR), temperature + dewpoint (°C), wind direction/speed/gust (kt), visibility, altimeter (inHg), and cloud layers. Source: NOAA Aviation Weather Center (keyless, public domain). |
| aviation.sigmet | Active in-flight weather hazard advisories (SIGMETs and AIRMETs) from the NOAA Aviation Weather Center. Returns current advisories over a lookback window (1-24h), optionally filtered by hazard (CONVECTIVE, TURB turbulence, ICE icing, IFR, MTN OBSCN mountain obscuration, ASH volcanic ash). Each advisory carries the issuing office, type (SIGMET/AIRMET/OUTLOOK), hazard, severity, valid-from/to times, affected altitude band (feet MSL), movement (direction/speed), and the raw text bulletin. Keyless, public-domain (NOAA). The active-hazard layer aviation agents need — distinct from aviation.metar (station conditions) and aviation.taf (forecasts). Not for navigation. |
| aviation.taf | Terminal Aerodrome Forecast (TAF) for one or more airports — the official aviation forecast issued for an airport (~24–30h ahead). Pass ids = comma-separated ICAO station ids (e.g. KATL). Returns the raw TAF text + issue time. Source: NOAA Aviation Weather Center (keyless). For current conditions use aviation.metar. |
| bank.lookup | FDIC-insured US bank directory lookup. Lookup by name (fuzzy match), FDIC certificate number, RSSD ID, or state. Each record includes name + short name, web address, active flag, city/state/county/zip, established + insured dates, charter class, bank class, primary federal regulator, number of branches, assets and deposits ($1000s), and closed-date + reason for inactive institutions. Public-domain FDIC data, free. |
| barcode.generate | Generate QR / Aztec / Data Matrix / PDF417 codes from structured payloads (url, wifi, vcard, vevent, email, sms, tel, geo, bitcoin, json, text). QR supports rounded or dotted modules, solid colors or linear/radial gradients, transparent backgrounds, configurable error correction, and a centered logo image (URL or data URI). |
| batch.run | Run up to 50 endpoint calls behind ONE x402 payment. Price is the exact sum of the sub-call prices (no batch discount). Atomic: every sub-call must succeed or nothing is charged — the failing calls are returned and you can retry for free. Eligible sub-calls are ordinary catalog endpoints (no bearer-only, deprecated, variable-priced, or metered-upstream endpoints, and no nested batch). |
| bio.gene | Look up a gene by its official symbol (e.g. "TP53", "BRCA1", "CFTR") and organism (taxid, default 9606 = human), merging NCBI Gene with UniProt. Returns the gene symbol, NCBI Gene ID, full description, organism, chromosome + cytogenetic map location, alias symbols, and the curated RefSeq function summary; plus the reviewed protein from UniProt — accession, entry id, recommended protein name, length in amino acids, and the curated function description. NCBI + UniProt links. Bioinformatics, clinical-genetics research, education. Public-domain NCBI + CC-BY UniProt data. Common organism taxids: human 9606, mouse 10090, zebrafish 7955, fruit fly 7227, yeast 559292. |
| bio.protein | Look up a protein in UniProtKB by accession (e.g. "P04637", "Q9Y6K9", "P0DTC2"). Returns the entry id, reviewed (Swiss-Prot) flag, recommended + alternative protein names, gene names, organism + taxid, sequence length and molecular weight, the curated function description, subcellular locations, Gene Ontology terms (id + term + aspect: biological_process / molecular_function / cellular_component), PDB structure ids, and keywords. Bioinformatics, structural biology, drug-target research, education. Gene-centric sibling: /api/bio/gene. CC-BY UniProt data. |
| bio.species | Resolve any organism by scientific or common name to the GBIF taxonomic backbone — the canonical free reference for life on Earth. Returns the accepted scientific name + canonical name, taxonomic rank and status, the match type/confidence, the full classification (kingdom → phylum → class → order → family → genus → species), up to 8 English vernacular (common) names, the count of recorded occurrences worldwide, and the GBIF species URL. Biodiversity, ecology, education, and as a taxonomy normalizer for other pipelines. Fuzzy-matches misspellings. CC-BY GBIF data. |
| bls.series | US Bureau of Labor Statistics time-series data. Pass seriesIds = comma-separated BLS series IDs (1-10 per call), optional startYear + endYear bracket (max 10 years). Each observation: year, period (M01-M12 monthly, Q01-Q04 quarterly, A01 annual, S01/S02 semiannual), periodName, value (verbatim string from BLS) + numericValue (parsed number), latest flag, footnote texts. Common series IDs: LNS14000000 (unemployment rate), CUUR0000SA0 (CPI-U all items), CES0000000001 (nonfarm employment), LNS11300000 (labor force participation). Unauthenticated upstream — ~25 queries/day per IP. |
| book.search | Open Library book metadata search. Lookup by free-text query (matches title + author), or by individual title / author / ISBN (10 or 13 digit). Each record includes Open Library work key, title, author names + Open Library author keys, first publish year, edition count, cover image URL, sample ISBNs, publishers, languages, subject tags, fulltext flag, ebook access tier (public/borrowable/printdisabled/no_ebook), and canonical openlibrary.org URL. CC0 public-domain metadata. |
| business.br-cnpj | Brazilian company registry lookup by CNPJ (the 14-digit national company tax ID). Returns the legal name (razão social), trade name, registration status and reason, start date, legal nature, company size, share capital (BRL), primary economic activity (CNAE code + description), contact email/phone, full address, and the partners/officers (QSA — name + role). Free, open Brazilian government data (Receita Federal via BrasilAPI). The canonical "who is this Brazilian company" lookup for KYB, diligence, and cross-border onboarding — complements business.lei (GLEIF) and business.sos-search (US). |
| business.entity-match | Fuzzy entity resolution: resolve a messy, free-text company name to its canonical GLEIF Legal Entity Identifier (LEI) with a 0-1 similarity score and a high/medium/low confidence label. Tolerant of legal-suffix noise (Inc/Ltd/GmbH/S.A.), word order, ampersands, punctuation, and former/alternate names (e.g. 'Apple Computer Inc' resolves to Apple Inc. via GLEIF's recorded other-names). Returns the top candidates ranked by score plus a single bestMatch (null when nothing clears medium confidence — so a low-confidence top hit is never silently treated as a match, which is the safe default for KYB). The record-linkage complement to business.lei (name search). Optional country (ISO-2) narrows to a jurisdiction. Backed by GLEIF Golden Copy (~2.6M entities, CC0). |
| business.entity-profile | Full business-entity dossier from a state registry — master record plus officers/principals, registered agent (name, phone, email, address), and filing history — the detail layer that flat registry search omits. v1 state: CT (Connecticut). Resolve by entityId (state registry record id), accountNumber, or name (partial; most recent registration wins). Returns entity status, type, registration date, mailing address, minority/woman/veteran/disability/LGBTQI ownership flags, officers, registered agent, and recent filings. KYB, vendor due-diligence, and counterparty-verification staple, sourced from the official state open-data portal. |
| business.entity-screen | KYC in one call: look up a business in a US state registry (NY, CO, CT) AND screen it against the OFAC sanctions list. Give a state and a name (or exact entityId). Returns the matched registered entities (id, type, status, jurisdiction, address, registered agent) and, for each, a sanctions screen of the entity name AND its registered agent against OFAC SDN — with fuzzy-match confidence and a flagged boolean. Counterparty due-diligence, vendor onboarding, AML. Composition of /api/business/sos-search + /api/law/sanctions-check (each section reports found/error independently). Note: sanctions screening is name-based and probabilistic — review flagged matches manually. |
| business.fi-companies | Official Finnish company registry search (PRH/YTJ avoindata, Finnish Patent & Registration Office). Search by company name. Each result: Business ID (Y-tunnus), current name, company form, trade-register status, primary line of business, registration date, and street address — descriptions in English where available. Free, CC BY 4.0. The authoritative FI company lookup — complements business.uk-companies and business.lei. |
| business.fr-companies | Official French company registry search (annuaire des entreprises / data.gouv.fr). Search by company name, SIREN, SIRET, or director. Each result: SIREN, legal name, legal-form code, primary NAF activity code, enterprise category (PME/ETI/GE), employee-count range, creation date, administrative status (active/ceased), head-office SIRET and address. Free, Licence Ouverte. The authoritative FR company lookup — complements business.uk-companies and business.lei. |
| business.kyb-360 | Full Know-Your-Business (KYB) intelligence dossier on a company, in one call. Pass name (legal/common company name); optional state narrows federal awards, and optional ticker pulls the company's SEC EDGAR identity + recent filings (SEC has no name search). Fans out to seven authoritative sources and merges them: SAM.gov registration (UEI/CAGE, active status), SAM exclusions (federal debarment/suspension), OFAC SDN sanctions screen, GLEIF LEI (legal-entity identifier + jurisdiction), USAspending federal contract awards, FARA foreign-agent registration (a disclosure status, not wrongdoing), and USPTO trademarks owned (brand/IP footprint). Returns headline riskFlags and a cleared boolean (strictly debarment + sanctions), a summary of every signal, and a found/error block per source so one slow or empty source never fails the rest. For vendor onboarding, KYB/AML, procurement due diligence, and competitive/IP research. Probabilistic name matching — confirm with a hard identifier (UEI/LEI/CIK) before acting; public records, not legal advice. |
| business.lei | Look up or search the global LEI (Legal Entity Identifier) registry — the authoritative ISO 17442 directory of ~2.6M legal entities worldwide. Pass `lei` for an exact 20-character LEI, or `query` to search by legal/other name (ranked best-match first). Filter name search by `country` (ISO 2-letter, HQ country) and `status` (active = currently active entities only; all = default). Each result returns the LEI, legal name, an alternate/other name, legal jurisdiction, entity category (GENERAL/FUND/BRANCH/…), ISO 20275 legal-form code, entity status (ACTIVE/INACTIVE), LEI registration status (ISSUED/LAPSED/RETIRED/…), headquarters city/region/country/postal code, the initial registration + last update + next renewal dates, and the managing LOU. Use this to canonicalize a company name to its LEI, resolve a counterparty, or enrich a vendor master. Data: GLEIF Golden Copy (CC0, public domain). |
| business.lei-hierarchy | Corporate ownership graph for a legal entity by LEI (GLEIF Level-2 relationships, live). Returns the direct parent and ultimate parent (each: LEI, legal name, jurisdiction, country, status), the direct children (paged), and total counts of direct and ultimate children. The authoritative 'who owns whom' lookup for KYB, beneficial-ownership, and corporate-tree mapping — complements business.lei (entity reference) and business.lei-isins. |
| business.lei-isins | ISIN ↔ LEI mapping (GLEIF, live, CC0). Two modes: pass lei to list every ISIN (security identifier) issued by that entity; or pass isin to resolve the issuer's LEI (with legal name, jurisdiction, country). Bridges securities to their legal-entity issuers for finance, compliance, and reference-data joins — complements business.lei and business.lei-hierarchy. |
| business.naics | Look up or search NAICS 2022 industry classification codes (US Census Bureau, public domain). Pass `code` for an exact NAICS code — 2–6 digits or a combined sector like 31-33 — and get its official title, hierarchy path, full official description, activity index terms, and direct child codes. Or pass `query` for free-text search (e.g. 'software publishers', 'coffee shop') over titles plus the official ~20k-entry activity index → ranked candidate codes, optionally filtered by `level` (2=sector … 6=national industry). Ground truth for industry coding in KYC, business registration, government filings, and ERP vendor/customer setup. |
| business.no-companies | Official Norwegian company registry search (Brønnøysund Enhetsregisteret). Search by company name. Each result: organisation number, name, organisation form, primary industry (NACE), employee count, registration date, website, bankruptcy and dissolution flags, and business address. Free, NLOD/CC BY 4.0. The authoritative NO company lookup — complements business.uk-companies and business.lei. |
| business.pl-krs | Official Polish company registry lookup by KRS number (KRS — Ministry of Justice, current extract / OdpisAktualny). Returns legal name, legal form, NIP and REGON identifiers, KRS registration date, share capital, and registered address. Register P = entrepreneurs (default); S = associations/foundations. Free, Polish public-sector open data. The authoritative PL company lookup — complements business.uk-companies and business.lei. |
| business.sos-search | Search state Secretary-of-State business registries, normalized to one schema across states. Currently supported: NY (active corporations + LLCs), CO (all entities incl. status), and CT (Business Registry Master incl. type, status, NAICS). Query by name (partial, case-insensitive) or exact entityId; results: state, entity id, name, type, status, formation jurisdiction, formation date, principal/registered address, registered agent. KYC, vendor due-diligence, and counterparty-verification staple. Each state sourced from its official open-data portal. |
| business.uk-companies | Official UK company registry (Companies House). Two modes: pass query to search companies by name (returns company number, name, status, type, incorporation date, address), or pass companyNumber for the full registry profile — legal name, status, company type, jurisdiction, incorporation/cessation dates, SIC activity codes, registered office address, accounts (next due, last made-up-to), confirmation-statement due date, charges and insolvency flags, previous names, and the officers list (name, role, appointed/resigned dates, nationality, occupation). Free, Open Government Licence (commercial use permitted). The authoritative UK-company KYB lookup — complements business.lei (GLEIF) and business.br-cnpj (Brazil). |
| calendar.business-days | Holiday-aware business-day calculator for 200+ countries. Three modes: pass start + addDays (signed integer) to shift a date by N business days; pass start + end to count business days between two dates (exclusive of start, inclusive of end); pass start alone to check one date (is it a business day, which holiday, next/previous business day). Query: country (ISO 3166-1 alpha-2), start (YYYY-MM-DD), optional region (subdivision code), addDays XOR end, optional weekend (comma-separated days, default sat,sun — e.g. fri,sat for the Gulf), optional types (holiday types treated as closures, default public,bank). Skipped holidays are itemized in the response. Use for payment terms, SLA deadlines, and delivery-date math. |
| calendar.earnings | Earnings release calendar — which US-listed companies report earnings in a date window, with expected and (once reported) actual EPS and revenue, and the time of day (before/after market). Pass a from/to window (YYYY-MM-DD; defaults to the next 14 days) and optionally a ticker to filter to one company. Agents cannot recall future earnings dates — this is the schedule. Data by Finnhub. |
| calendar.holidays | List the official holidays for a country and year, with exact observed dates including substitute days (e.g. a Saturday July 4th observed on Friday). Query: country (ISO 3166-1 alpha-2, 200+ supported), year, optional region (subdivision code like CA for US-California or BY for DE-Bavaria), optional types filter (public, bank, school, optional, observance — comma-separated), optional lang (ISO 639-1) for localized names. Returns one item per holiday: { date (YYYY-MM-DD), name, type, substitute, rule }. Dates are computed from maintained per-country rules — including movable feasts and lunar-calendar holidays — so they stay correct year over year. Use for scheduling, delivery estimates, payment terms, and SLA math. |
| calendar.ipo | IPO calendar — companies going public (or recently public) in a date window, with the expected date, symbol, name, exchange, price range, number of shares, total offering value, and status (expected/priced/filed/withdrawn). Pass a from/to window (YYYY-MM-DD; defaults to a ±30-day span around today). Data by Finnhub. |
| census.demographics | US Census ACS 5-year demographics for a state or county. Give a state (2-letter or 2-digit FIPS), optionally a 3-digit county FIPS, and get population, median age, median household income, per-capita income, poverty rate (computed), households, owner-occupancy rate (computed), median home value, and median gross rent. Free, public-domain (US Census Bureau). Complements census.zipcode (ZIP/ZCTA-level) with state- and county-level geography. Authoritative demographic data an LLM cannot recall — for market sizing, site selection, and socioeconomic research. |