TariffMonkee

Pay-per-call web data APIs for AI agents over the x402 protocol. Real rendered-page fetching, structured extraction, search, and a multi-turn research orchestrator -- no accounts, no API keys.

Skills

• Verify Citation
  Use this when you have a claim and a source cited for it and need to know whether that source actually says what it is cited for -- checking your own citations before publishing, auditing someone else's document, or validating references in AI-generated text. Returns supported, contradicted, or not_addressed, always with the exact verbatim passage from the source as evidence, so the verdict can be checked rather than trusted. A dead, paywalled, or PDF-only source returns source_unavailable rather than a false contradicted, and a page stating it has been retracted is flagged. Accepts a URL or a bare DOI. For a whole document use the batch form, which dedupes repeated sources so many claims against one paper cost one fetch.
  api
• Crawler Diligence
  Use this when assessing whether a website is getting AI crawler traffic it is not charging for -- due diligence before buying a content site, or auditing your own. Screening mode needs only a domain and probes from outside: per-crawler robots.txt rules for 18 named AI crawlers, and whether access is already sold through TollBit, RSL, x402, or a 402 returned to a declared crawler. Verified mode takes real server logs you already have and breaks traffic down by crawler and operator, using reverse DNS to flag IPs that claim a crawler User-Agent but do not belong to that operator. No revenue figure is invented: supply your own per-crawl rate and one is computed as a band, or omit it and get counts and posture only.
  api
• Tw Reference
  Use this when you need Taiwan government immigration or work-permit rules in English -- ARC and visa categories, the Gold Card, APRC, and all 15 WDA work-permit categories (A01 through A15). Call with no parameters first to get the index of available categories and ids, then call again with a category to get the full record: labelled sections covering qualifications, required documents, application procedure, and Q&A, translated from the original Chinese. This returns SOURCE REFERENCE DATA, not an eligibility decision -- it tells you what the published rules say, it does not judge whether a specific person qualifies. Every response carries the source_url and a translation notice; where the translation and the original disagree, the Chinese original is authoritative. Check section_count on a record before relying on it for completeness.
  api
• Content
  Use this when you need the readable text of a specific web page and either don't have your own scraper, or the page needs real JavaScript execution to render correctly (single-page apps, client-side-rendered docs, dynamically loaded articles) -- a plain HTTP fetch of the raw HTML would miss that content. Requires knowing the URL already; use tariffmonkee-search first if you don't. Returns Markdown, not structured data -- pair with tariffmonkee-extract if you need specific fields pulled out of the result rather than the whole page. Repeat calls for the same URL within 24 hours are served from cache, so re-checking a page you already fetched is cheap.
  contentbrowsermarkdownscraping
• Screenshot
  Use this when you need a visual capture of a page, not its text -- confirming what a page actually looks like, checking that a UI or layout change took visible effect, or producing visual evidence of a page's state at a point in time. The HTTP response body is the raw PNG image itself, not JSON. Renders with real JavaScript execution first, same as tariffmonkee-content, so client-side-rendered pages capture correctly rather than showing a blank or loading state.
  screenshotbrowservisualpng
• Pdf
  Use this when you need an archival or printable snapshot of a full page as a file -- a record of a page's content and layout at a point in time, suitable for saving, sharing, or printing. The HTTP response body is the raw PDF file itself, not JSON. If you only need the text content and not a file, tariffmonkee-content is cheaper and returns Markdown directly.
  pdfbrowserarchival
• Extract
  Use this when you already have raw text or Markdown content in hand -- from a prior tariffmonkee-content call, or text already in your own context -- and want specific structured fields pulled out of it against a JSON Schema you supply. Pure AI extraction with no browser render involved, so it's cheaper and faster than tariffmonkee-extract-url. Use tariffmonkee-extract-url instead if you only have a URL and haven't fetched the page content yet.
  extractionstructured-dataai
• Extract Url
  Use this when you have a URL but not the page content yet, and want structured data extracted from it in a single call -- fetches with a real rendered browser and extracts against your JSON Schema in one round trip, trying multiple AI models in sequence for reliability. If you already have the content in hand, tariffmonkee-extract is cheaper since it skips the browser render entirely. Results are cached for 24 hours per URL+schema combination, so repeating the same extraction on the same page is cheap.
  extractionstructured-dataaibrowser
• Aeo Audit
  Use this when you want to know how ready a page is for AI and agent consumption specifically, not human browsing -- it checks schema.org structured data, crawlability (robots.txt, meta robots, canonical tags), and llms.txt presence, then returns a 0-100 score with specific, actionable fixes. Useful for auditing your own site, a client's site, or deciding how much to trust a page as a citable source.
  auditseoaeoschema.org
• Search
  Use this when you need to find pages on a topic and don't already know a specific URL -- this is the entry point, not tariffmonkee-content, when you're starting from a question rather than a known page. Backed by three search providers with automatic failover, so a single provider outage doesn't take it down. Pass region=tw for a Traditional Chinese / Taiwan-region search lane instead of the global default. Returns titles, URLs, and snippets only -- pair with tariffmonkee-content for full page text, or use tariffmonkee-research instead if you want both search and fetch done for you in one call.
  searchdiscovery
• Research
  Use this instead of chaining tariffmonkee-search followed by several tariffmonkee-content calls yourself -- it does both in one call, deduping sources by domain and packing up to 10 rendered sources into a single roughly-8,000-token bundle sized for direct use in your context. Requires at least 3 successfully fetched sources to succeed, and reports sources_requested vs. sources_returned so you can see how many of the candidate set actually came back. Costs more than a single search or content call, but replaces several of them -- reach for this when you want a research bundle for a topic, not a single page or a single search results list.
  api
• Linkcheck
  Use this when you need a safety verdict on a specific URL before fetching it, sharing it, or acting on it -- checks the redirect chain, detects homograph/punycode hostname tricks, and cross-references the destination against the URLhaus threat feed. Not a domain-wide audit; use tariffmonkee-sitegrade for configuration quality or tariffmonkee-domain-intel for ownership facts. If you need all three synthesized together, use tariffmonkee-trust-audit instead.
  securityurlreputationthreat-intel
• Sitegrade
  Use this when you need to grade how well a domain is technically configured -- HTTPS reachability, security headers (HSTS, CSP, X-Frame-Options, and more), and CAA records, returned as a 0-100 score with an A-F letter grade. Reports HTTPS connect success/failure, not raw certificate expiry -- Cloudflare Workers can't read a remote server's certificate metadata directly. Not for ownership/registration facts (use tariffmonkee-domain-intel) or single-URL reputation (use tariffmonkee-linkcheck).
  securityheaderstlsaudit
• Domain Intel
  Use this when you need registration and ownership facts for a domain -- age, registrar, privacy-proxy status, nameservers, and mail provider, via RDAP. A domain younger than 30 days is flagged directly in the result. Not a configuration grade (use tariffmonkee-sitegrade) or URL-specific reputation check (use tariffmonkee-linkcheck).
  securitywhoisrdapdomain
• Email Verify
  Use this when you need a deliverability verdict on a single email address -- syntax, MX record presence, disposable-domain match, and SPF/DMARC presence on its domain. Scoped to one address, not a domain-wide audit; use tariffmonkee-sitegrade if you need the sending domain's broader security posture.
  securityemaildeliverabilityverification
• Trust Audit
  Use this instead of calling tariffmonkee-linkcheck, tariffmonkee-sitegrade, and tariffmonkee-domain-intel separately when you want one synthesized trust read -- it runs all three (plus tariffmonkee-email-verify if you supply an email) and returns a confidence-weighted verdict that explicitly flags disagreement between sources, e.g. a brand-new domain paired with an unusually mature TLS/header setup, rather than just concatenating four results. Costs more than any single check, but replaces several of them -- reach for this when you want the full picture, not one signal. Supply at least one of domain, url, or email.
  securityauditorchestrationsynthesis
• Gbp Audit
  Use this when you need to audit a business's Google Business Profile -- missing fields (phone, website, description, category, photos, hours), claimed status, and review health (rating, review count, low-rating share). Requires a location string, not just a business name, since common business names are ambiguous without one. AI-readiness in the result reuses tariffmonkee-aeo-audit against the listing's own domain when one exists.
  local-seogoogle-business-profileauditreviews
• Serp Check
  Use this when you need to know whether and where a specific domain ranks for a specific keyword in Google's organic results -- not a list of top results (use tariffmonkee-search for that), a targeted rank check for one domain you already have in mind.
  local-seorank-trackingserp
• Ai Visibility
  Use this when you need to know whether a business actually gets mentioned or cited when a real AI model with web search answers a specific question -- sends the real query to a live model and checks the real response, not a prediction. Costs more than most flat endpoints since it pays for an underlying AI model call; use it when the specific query matters, not for broad monitoring.
  local-seogeoai-searchbrand-monitoring
• Visibility Report
  Use this when you want one combined, white-label local visibility report for a business rather than calling gbp-audit, serp-check, and ai-visibility separately -- it runs all three and returns a single self-contained HTML report with no TariffMonkee branding, suitable for an agency to present under its own name. Costs more than any single check, but replaces three of them and adds report formatting.
  local-seoreportingwhite-labelsynthesis
• Lead List
  Use this when you have a list of candidate local businesses (up to 10 per call) and want to know which ones are actually worth pitching -- each candidate is checked against its Google Business Profile and only businesses with at least one real red flag (unclaimed, missing photos, missing hours, missing description, missing website, low rating, or few reviews) are returned, as a CSV ready for cold outreach. Cheaper per-business than calling gbp-audit individually across a full batch.
  local-seoprospectinglead-gencsv
• Work Permit Check
  Use this when you need eligibility GUIDANCE (not a determination) for a foreign professional's Taiwan work permit, given a profile of education, years of experience, offered salary, and a few yes/no flags. Checks the 2026 graduate Extended Stay exemption first (no work permit needed at all, the highest-value answer this returns), then the general professional route (an Article 5 qualification route plus the NT$47,971 salary floor), and reports the points-route criteria and confirmed scoring bands as attributed reference WITHOUT computing a total, because the official per-band table could not be verified from an authoritative source. Only Taiwan's Ministry of Labor can decide an actual application, and the employer files it, not the individual. Pair with tariffmonkee-tw-reference for the underlying category text.
  api
• Tax Residency Check
  Use this when you need Taiwan tax residency GUIDANCE (not tax advice) for a specific person and year -- given their Taiwan arrival/departure dates and income, it determines resident (183+ days in the calendar year) vs non-resident status using the official day-counting rule, then estimates what's actually owed: for non-residents, the 6%/18% flat withholding rate; for residents, an estimated tax bill using that calendar year's verified brackets and deductions, compared against tax already withheld if supplied. Flags the 90-day offshore-employer anti-avoidance rule and mid-year status changes as informational notes. Does not compute DTA treaty rates or the Alternative Minimum Tax -- both are flagged, not calculated, when the inputs suggest they might apply. Pair with tariffmonkee-work-permit-check for the immigration side of a Taiwan move.
  api
• Healthcare Navigator
  Use this when you need to route someone to NHI-contracted care in Taiwan and explain what it will cost -- given a county and optionally a specialty keyword, it searches NHIA's live open-data roster of ~21,500 contracted institutions; given an NHI eligibility profile (start date, basis such as employed vs. dependent vs. self-insured, and any trips abroad), it computes whether the 6-month wait applies and, if so, the exact date it clears. Always returns the current NHIA copay schedule by facility tier and referral status. Explicitly not medical advice -- no diagnosis, no doctor recommendation -- and never reports English-language capability at a clinic, since that isn't an official data field. Pair with tariffmonkee-work-permit-check and tariffmonkee-tax-residency-check for the other sides of a Taiwan move.
  api
• Tw Rental Check
  Use this when someone has a Taiwan rental listing or draft lease and wants to know if anything in it is illegal or a known scam pattern before signing -- pass the listing text or a URL and it checks the deposit against the legal 2-month cap, the electricity rate against a verified post-2024 billing rule, whether the lease tries to (unenforceably) ban household registration or a tax deduction claim, and flags rooftop-addition mentions and pre-viewing-deposit requests. Guidance and legal exposure, not a legal determination -- pair with tariffmonkee-work-permit-check, tariffmonkee-tax-residency-check, and tariffmonkee-healthcare-navigator for the other sides of a Taiwan move.
  api
• Tw Property Check
  Use this when someone is considering buying a specific Taiwan property and wants a reality check before making an offer -- given the address, it compares the asking price against real recent sale-price comparables in the same district, gives construction-year context against Taiwan's seismic code history (correctly noting this is not a single 1999 cutoff), and flags rooftop-addition/unregistered-structure mentions if listing text is supplied. Per-address liquefaction risk is explicitly not available -- flagged honestly with a link to the official map rather than guessed at. Pair with tariffmonkee-work-permit-check, tariffmonkee-tax-residency-check, tariffmonkee-healthcare-navigator, and tariffmonkee-tw-rental-check for the other sides of a Taiwan move.
  api
• Lead Discovery
  Use this when you don't already have a list of candidate businesses and want to discover them by category and area -- searches Google Maps for the given category in the given area and returns only businesses with at least one real red flag (unclaimed, missing photos, missing hours, missing website, low rating, or few reviews), as a CSV ready for cold outreach. Use tariffmonkee-lead-list instead if you already have specific businesses in mind.
  local-seoprospectinglead-gendiscoverycsv
• Tw Localize Check
  Use this when you have Traditional Chinese text and need to know if it reads as mainland (PRC) or Hong Kong wording instead of genuine Taiwan usage -- checking your own translation before publishing to a Taiwan audience, auditing localization work, or cleaning up AI-generated Traditional Chinese that defaulted to mainland vocabulary. This is a judgment layer on top of simplified-to-traditional character conversion, not a substitute for it: a converter can turn 软件 into 軟件 with perfect grammar and it will still read as non-Taiwanese. Each flag names the term, the Taiwan-correct replacement, category, severity, and why, and a few terms carry a context_note where the flagged wording is also legitimate Taiwan usage in a different context. Pass fix=true to also get a corrected version of the text. Backed by a curated, growing glossary (47 terms at launch, not a full linguistic model) -- it will miss terms not yet catalogued.
  api
• Tw Fraud Check
  Use this to check whether a domain or URL is a known Taiwan fraud site (checked against NPA's DNS-suspension blocklist and a separate report-volume dataset with report counts and dates), or to check whether a claim/rumor matches 165's archive of officially debunked scam stories. Provide 'domain' or 'url' for the site check, 'claim' for the rumor check, or both together. A domain not found in either dataset is NOT confirmed safe -- it may simply not have been reported yet. A rumor with no match is NOT confirmed true -- the rumor archive's last entry is from April 2025, a disclosed gap, not a claim of full coverage. Informational only; NPA's 165 system is the authoritative source.
  api
• Tw Business Day Check
  Use this when you need to know whether a specific date is a Taiwan government office business day -- weekends and public holidays both count as non-business days, matching the ROC government's own official office calendar rather than a fixed list of holiday dates. Pass 'range_end' in addition to 'date' to also get a business-day count and the list of holidays across a range, useful for computing a real turnaround time like '5 business days from now'. Coverage runs from roughly 2017 onward and the following year's data appears around June each year; a date outside that range returns a clear error rather than a guess. The 'description' field for each day is passed through as-is from the source and is worded inconsistently across years -- only 'is_holiday'/'is_business_day' should be used for logic.
  api
• Tw Transit Check
  Use this when you need real Taiwan transit data right now -- bus arrival predictions for a specific route, TRA (regular rail) live board for a station, THSR (high speed rail) today's train info, or MRT arrival boards for Taipei/Kaohsiung/Taoyuan/Kaohsiung Light Rail. This is a live data pass-through, not a trip planner -- it does not compute a route or transfer plan, it reports what TDX (Taiwan's official transit data platform) says right now for the specific stop/route/station you ask about. Pass mode plus the relevant identifiers: city+route_name for bus, station_id for tra/thsr, rail_system (+optional station_name) for mrt. Bus responses are curated and handle a real TDX quirk (a -1 estimate means the bus already passed, not missing data); tra/thsr/mrt responses return TDX's own fields as-is, disclosed as not yet remapped. MRT here means arrival boards only, not precise train position -- Taipei Metro's GPS position feed needs a separate application directly with Taipei Rapid Transit Corporation and isn't covered.
  api
• Research Orchestrator
  Multi-turn research: checks your query is specific enough to search, automatically broadens and retries once if the first pass returns too few sources, and asks a clarifying question -- pausing the task, resumable with a follow-up message carrying the same taskId -- only if that isn't enough. Wraps the same research engine as tariffmonkee-research, with real iterative reasoning around it rather than a single fixed call.
  researchsearchorchestrationmulti-turnagentic

How to call

https://tariffmonkee.com/paid/a2a

Listed in

Directories this entry was found in.