All Services

Fetches a web page by URL and returns the readable article content as clean Markdown, stripping navigation, ads, and boilerplate.

Returns an avoid, caution, or clear risk verdict for a token contract address on Solana or Base before an agent executes a trade

Analyzes a public GitHub repository deterministically and generates an AGENTS.md file with quoted commands and verified file paths from the repository's manifest and git tree.

Deterministically analyzes a public GitHub repository and reports commands, file health, phantom paths, and primary language without using any language model

Verifies a single legal citation against a 14.2M-case corpus (CourtListener), detecting whether the case name matches the reporter citation or reveals a hallucinated reference.

Returns RGE (Reconnu Garant de l'Environnement) energy-renovation certifications for a French establishment identified by SIRET, including qualification details, validity, domains, and certifying body.

Executes submitted code and test assertions in an isolated sandbox and returns a pass/fail verdict with stdout, stderr, and exit code evidence.

Fetches a URL server-side and checks whether a specific claim or text actually appears on that page, returning match details and confidence.

Verifies whether a USDC transaction hash settled on-chain and returns the transfers it contains, with optional recipient and minimum-amount checks

Retrieves vesting schedule and equity grant data for a specified person from Aviato's intelligence platform.

Returns validated, multi-source web evidence packs answering a research query, with source citations, freshness metadata, and validation status.

Retrieves TikTok videos that use a specific song/audio track

Look up metadata for an aerial view video by its video ID, returning details such as status, duration, and availability

Decodes a 17-character Vehicle Identification Number (VIN) into detailed vehicle specifications including make, model, year, engine, transmission, and more using NHTSA vPIC data.

Decodes a 17-character VIN into structured vehicle details including make, model, year, engine, transmission, body class, and assembly plant, with check-digit validation.

Returns active safety recalls for a vehicle (US and Canada) given its 17-character VIN

Adds paid runtime hours to an existing anonymous metered VM, extending its active lifetime

Provisions an anonymous, metered virtual machine paid per-hour via Lightning (L402) or USDC on Base (x402), returning a one-time bearer token for access

Clone a voice from an audio sample to produce a reusable voice handle for speech synthesis on Venice AI's TTS models.

Solves allocation, assignment, selection, and resource-mix problems to find the lowest-cost or highest-value feasible solution given caller-supplied options, capacities, budgets, and eligibility rules.

Scores and audits an OpenAPI description across six dimensions to determine how well-suited it is for autonomous AI agent consumption.

Looks up a CVE by ID and returns its EPSS exploit probability score, severity ratings, and a plain-language summary

Returns a per-token profit-and-loss summary (realized and unrealized PnL, buy/sell volume and trade counts) for a given Solana wallet address over a chosen time resolution.

Schedule a single future HTTPS callback to a given URL at a specified UTC time, paid per-use with USDC via x402.

Returns the native ETH balance plus USDC, USDT (Ethereum), and up to 20 additional ERC-20 tokens for a given wallet or contract address on Base, Ethereum, or Solana, read live from the chain.

Returns a 0-100 risk score, a verdict, and explained risk flags for any Base blockchain address based on its x402 payment history

Returns a deterministic 0-100 trust score with itemized sub-scores for any Base wallet or contract, computed from on-chain activity, funds, identity signals, and explorer reputation flags.

Returns a deterministic 0-100 trust score for any Base wallet or contract address, based on on-chain activity, identity signals, and scam flags

Returns the most recent AIS position fix for a single vessel identified by IMO number, including lat/lon, speed, course, destination string, and data age.

Fetches the HTTP response headers for a given URL, returning status, redirect info, timing, and all headers without downloading the full body.

Fetches, normalizes, hashes, and returns the content of any URL so an agent can detect changes between calls using the returned content hash.

Fetches two versions of a URL's content, normalizes and hashes them, then returns a structured diff showing exactly what changed between the before and after snapshots.

Screens a person, company, vessel, or aircraft name against 10+ global sanctions and export-control lists using fuzzy, typo-tolerant matching against versioned list snapshots.

Returns all active NWS watch, warning, and advisory polygons that contain a given latitude/longitude point.

Finds the best-matching decision-maker at a target company by searching through a prioritized cascade of job title tiers, returning the top people from the highest-priority tier with results.

Returns per-hour multi-model ensemble statistics (median, p15/p85 band, min/max, model count, agreement score, rain probability, frost assessment) plus active DWD warnings for a given lat/lon point.

Returns per-hour ensemble statistics (median, p15/p85 band, min/max, model count, agreement score, rain probability, frost assessment) plus DWD warnings for a geographic point, pooled across up to 12 NWP models.

Returns 7-day hourly and daily weather forecasts for any lat/lon coordinate, including temperature, precipitation, humidity, and wind speed

Returns real-time weather conditions (temperature, humidity, wind, UV index, pressure, precipitation) for a given latitude/longitude coordinate

Fetches any public web page or PDF by URL and returns clean, LLM-ready Markdown text with title and metadata, billed per-call via x402.

Renders any public URL in a real browser and returns a PNG or JPEG screenshot as base64 JSON or raw image bytes

Runs seven simultaneous page analyses (meta SEO, alt-text, social cards, internal links, WCAG accessibility, structured data, robots/llms.txt) on a single URL and returns a scored 0-100 report with itemized deductions.

Audits the <head> section of any public webpage for on-page SEO fundamentals including title, meta description, robots, canonical, hreflang, OG/Twitter tags, and more

Detects the CMS, framework, or site-builder powering any public URL using deterministic HTTP-level fingerprinting, returning the platform name, category, confidence, evidence, and server/CDN stack.

Generates a complete brand identity bundle — logo (SVG+PNG), app icon, social card, and WCAG-checked color palette — in a single API call.

Checks a website's robots.txt and llms.txt files, returning crawler access verdicts for major search and AI bots, declared sitemaps, syntax warnings, and llms.txt structure summary.

Generates a finished brand logo (SVG + PNG) from a company name, icon, colors, shape, and layout preferences

Audits any public webpage for image accessibility issues including missing alt attributes, filename-as-alt, generic/over-long alt text, unlabeled SVGs, image-map areas, and image inputs without alternatives.

Calculates the WCAG contrast ratio between a foreground and background color, returning the exact ratio and AA/AAA pass/fail verdicts for normal text, large text, and UI components.

Performs a single socket-level HEAD probe of a URL and returns a full timing waterfall (DNS/TCP/TLS/TTFB), TLS certificate details, HTTP/2 support, hosting/CDN info, redirect chain, and header analysis.

Evaluates a math expression at 50-significant-digit BigNumber precision and returns the exact result as a string

Generates a scannable QR code from any text or URL, returning SVG markup, PNG as base64, and a data URI in one response.

Performs static WCAG accessibility analysis on a public URL, mapping findings to A/AA/AAA success criteria across alt text, headings, ARIA, forms, and more

Extracts structured data (links, headings, markdown, images, metadata) from one or up to 50 URLs in a single pay-per-call request

Captures a screenshot of any URL as PNG, JPEG, or PDF, with options for full-page capture, mobile viewport, post-load actions, and structured content extraction.

Fetches and extracts the readable content of a web page as Markdown, with integrity hash, for use as verifiable evidence

Follows a URL's full redirect chain (301/302/303/307/308) and returns every hop with its HTTP status code and Location header, revealing the final destination of shortened or cloaked links.

Discovers and parses a website's XML sitemaps (via robots.txt, explicit URL, or /sitemap.xml) into a bounded list of URLs with metadata including lastmod, changefreq, and priority

Fetches a URL's HTTP response headers and audits the presence of key security headers (HSTS, CSP, X-Frame-Options, etc.) without downloading the response body.

Creates a signed, timestamped cryptographic proof of a web page's content — including rendered text, screenshot, TLS certificate, and serving IP — that anyone can verify later without trusting the service.

Fetches a web page and returns an AI-generated summary with optional topic focus, length control, and language selection

Fetches a public web page and returns its main content as clean Markdown, stripping menus, ads, and boilerplate, plus metadata like title, byline, publish date, and word count.

Pay-per-call web search endpoint that runs one or more queries and returns ranked results with optional page extracts and synthesized summaries.

Searches the web or scrapes a URL and returns clean, LLM-ready markdown with source URLs, powered by Jina Reader

Performs a web search and returns ranked results with title, URL, snippet, publish date, and up to 1500 characters of cleaned main page text fetched live from each result.

Fetches a public URL and returns its content as Markdown (or plain text for JS-heavy pages), with optional link extraction.

Performs a pay-per-call single-page technical SEO and website quality audit, returning a graded report with score, issues, and category breakdowns.

Parses a public URL into structured fields: main text, headings, links, tables, and rich metadata (Open Graph, Twitter cards, JSON-LD, canonical, lang) without a browser.

Fetches a web page by URL or converts raw HTML into clean, LLM-ready Markdown with structured metadata (title, description, OG tags, canonical URL, language)

Decodes EVM signature requests (EIP-712, Permit/Permit2, ERC-20 approve calldata) into structured human-readable intent with token, spender, amount, expiry, risk flags, and action hints

Returns the top buyers and sellers of any ERC-20 token in the last hour on Base, Ethereum, Arbitrum, Optimism, or Polygon, with wallet-level net USD flow, trade counts, smart-money labels, and bot flags.

Analyzes DEX swap events for a token over 1–24 hours to return buy/sell counts, USD volumes, net flow, top wallets, and largest trades across 7 major EVM chains.

Returns a paginated list of companies using a specified technology, with optional firmographic filters and enrichment, billed per record returned.

Returns detailed sommelier guidance for a named wine including grape variety sensory profiles, terroir/appellation data, producer heritage, and serving/decanting recommendations.

Returns investment-grade wine market metrics including price index, financial returns across multiple timeframes, liquidity, volatility, Sharpe ratio, bid signals, price spread, and drinking maturity window for a named wine.

Purchase API credits for the witnessd neutral-witness service by paying $1 USDC, receiving an API key and credit balance in return.

Answers natural language queries as a simplified image response using Wolfram|Alpha's computational knowledge engine, with customizable display options.

Fetches a World Bank economic indicator time series (e.g. GDP, inflation, unemployment, population) for a given country and year range

Compares a single World Bank development indicator across up to 10 countries simultaneously, returning the most recent value per country sorted descending.

Check whether a given phone number is registered on WhatsApp and retrieve basic account status

Look up carrier, line type, country, and location details for any phone number

Converts a plain-text prompt into clean, layered SVG code optimized for icons, charts, patterns, and flat backgrounds in about 8 seconds.

Returns the implied-volatility surface (expiry × strike grid with IV, delta, and gamma) for a given crypto asset from Deribit options data.

Returns order-book depth at 50, 100, and 200 basis points per crypto asset — bid/ask USD notional and imbalance ratio — aggregated from Binance spot books and refreshed every 15 minutes.

Decomposes an asset's composite score into per-factor-class contributions and weights at a given forward return horizon, explaining why the asset is ranked where it is.

Returns a daily AI-readable crypto market brief with top-ranked assets by composite score, total signal triggers, anchor day, and a structured narrative summary for LLM grounding.

Returns highest- and lowest-conviction crypto assets ranked by composite score, with the count of atomic signals contributing per asset, for filtered watchlists and daily briefings.

Returns a ranked leaderboard of quantitative trading signals with IC, hit rate, IR, grade, and asset coverage evaluated at a chosen forward return horizon.

Returns the top predictive metrics for a single crypto asset ranked by information coefficient (IC) and hit rate, identifying the most reliable signal inputs for that specific asset.

Backtests a custom-weighted composite across 8 factor classes, returning per-day composite return series with Sharpe, Sortino, and max drawdown metrics.

Returns the full registry of all 143 atomic signals and 9 IC-weighted composite signals with metadata including parameters, 30-day trigger statistics, geometry type, factor class, and category — the canonical capability-discovery call for the CRYPTYX platform.

Returns macro regime classification (expansion, consolidation, or contraction) with a confidence score for a given crypto asset, derived from factor signals rather than price action.

Returns audit-grade explainability for a single signal×asset×day combination, including full trigger reason, factor scores by class, composite score, parameters in effect, and regime context.

Returns the peer cluster label and similarity-ranked peer assets for a given crypto asset, computed from rolling 30-day factor correlations.

Returns all currently-triggered trading signals for a single crypto asset, with confidence scores, grades, and trigger reasons.

Returns a complete per-asset trading thesis including composite score, rank, dominant factor class with t-score, top firing signals, macro regime, and key metrics in a single call.

Returns the current perpetual futures funding rate for a crypto asset, including annualized rate and z-score vs trailing 30-day distribution, sourced from Deribit and CoinGlass.

Returns t-scores for a specific asset across CRYPTYX's 8 canonical factor classes (CORR, EFF, FLOW, FUT, OB, OPT, TR, VOL) at a given forward return horizon (7d, 14d, or 30d).

Returns a time series of signal firing events for a single crypto asset over the last N days, including signal ID, date, trigger status, and confidence.

Returns today's live firing signals across ~200 crypto assets, including atomic and composite signals with z-scores, confidence, and macro regime context.