145 Services from x402-deployer.x402-deployer.workers.dev

Generates a safe-for-work image from a text prompt using one of four Venice-backed AI models, returning a permanent fal-hosted PNG URL within 5 seconds for $0.01 USDC.

Scans text for personally identifiable information (emails, phones, SSNs, credit cards, addresses, names, IPs, API tokens), returns matches with risk level, and optionally redacts them with [TYPE] placeholder tokens.

Upscales an image 2x or 4x using AI super-resolution via Venice's image/upscale endpoint, returning a permanent fal-hosted URL

Analyzes text for overall sentiment (score -1 to +1), per-emotion breakdown (joy, anger, sadness, fear, surprise, disgust), and optional aspect-based sentiment

Converts text to speech audio using multiple TTS providers (Kokoro, ElevenLabs, Orpheus, MiniMax) with 30+ voices and multiple output formats.

Classifies text into caller-supplied labels using zero-shot inference, supporting multi-label mode with calibrated confidence scores and reasoning.

Classifies arbitrary text into caller-supplied categories using zero-shot classification with calibrated confidence scores and reasoning, powered by Mistral.

Fetches verified source code, ABI, compiler settings, and metadata for any deployed smart contract on Ethereum, Base, Optimism, Arbitrum, or Polygon.

Redacts PII (emails, phones, SSNs, IBANs, credit cards, IPs, names, addresses, dates of birth) from text using a two-stage regex + LLM pipeline, returning sanitized text and a list of every redacted entity.

Fetches an HTML page and returns every anchor link with its text, rel attribute, and whether it is external, up to a 500-link cap

Queries IANA-mandated RDAP to return full domain registration details including age, expiry, registrar, nameservers, DNSSEC status, EPP statuses, risk flags, and optionally certificate-transparency summary.

Looks up live crypto token prices (USD + native currency), 24h/7d change, market cap, volume, supply, ATH, and ATL by token ID, symbol, or contract address with chain.

Fetches live CFTC-regulated prediction market contracts from Kalshi, with prices normalized to 0–1 probabilities, filterable by status, series, event, query, and volume.

Parses a 5/6/7-field cron expression or named macro and returns a plain-English explanation, frequency bucket, validation status, and runs-per-day estimate

Simulates buy/sell/transfer transactions for any Base or Ethereum ERC-20 token and returns a honeypot verdict, tax percentages, and reason string to detect sell-blocking scams before swapping.

Fetches paper metadata from the arXiv API and generates a properly-formatted BibTeX citation entry with an auto-derived or custom cite key.

Fetches a URL, strips HTML to plain text, hashes it with SHA-256, and returns a changed flag plus added/removed line diffs compared to a previous hash or text, with optional CSS selector scoping.

Scores a resume against a job description on a 0-100 fit scale with calibrated subscores for keyword match, experience, skills, formatting, and impact, plus ranked improvement suggestions.

Converts a natural language prompt into a SQL query for a specified database dialect, returning the SQL, an explanation, referenced tables, destructiveness flag, and warnings.

Returns sunrise, sunset, solar noon, day length, golden hour, dawn/dusk, and civil/nautical/astronomical twilight times for any latitude/longitude on any date

Detects homoglyph and lookalike characters (Cyrillic, Greek, fullwidth, mathematical) used in phishing attacks, plus zero-width and RTL-override characters in text strings.

Parses a User-Agent string into browser, OS, device type, and bot/crawler identity using pure local regex — no upstream calls

Returns the current star count and a time-series of cumulative stars vs. date for any public GitHub repository, sampled across up to 10 pages of stargazers.

Given a passport country and destination country, returns visa entry type (visa-free, e-visa, visa-on-arrival, or visa-required), permitted stay in days, and a human-readable status string; supports full 200x200 visa matrix retrieval.

Returns current weather conditions plus 1–7 day daily and optional hourly forecasts for any latitude/longitude coordinate, powered by Open-Meteo.

Computes sunrise, sunset, solar noon, day length, twilight times (civil/nautical/astronomical), golden hour, dawn, and dusk for any latitude/longitude on any date, returning UTC ISO8601 strings.

Computes a 64-bit SimHash fingerprint over token-level k-shingles (default k=3) for near-duplicate content detection, returning hex and decimal forms plus token and shingle counts.

Resolves an IPv4 or IPv6 address to ASN number, ASN org, ISP, country, region, city, postal code, lat/lng, and timezone

Checks what percentage of an LP token's supply is locked in Unicrypt v2, Team Finance, or burn addresses and returns a rug-pull risk verdict (locked/partial/none)

Looks up airport details by IATA or ICAO code, returning name, type, location coordinates, country, region, municipality, scheduled service flag, and reference links.

Returns structured financial line items (revenue, net income, EPS, total assets, cash, etc.) from SEC XBRL filings for any US-listed ticker

Generates complete email drafts (subject, body, salutation, sign-off) across 10 email types and 7 tones with length controls

Transcribes a video or audio URL using Whisper v3 and returns a downloadable SRT or VTT subtitle file.

Audits a domain's email authentication posture by reading its DNS TXT records for SPF, DMARC, and DKIM, then returns a 0-100 score, A-F grade, and human-readable findings explaining spoofing resistance and deliverability readiness.

Fetches a source URL and uses an LLM to determine whether a given claim is supported, contradicted, or absent in that source, returning a verdict, confidence score, and evidence quote.

Analyzes password strength via entropy, character classes, and crack-time estimates, then checks if it has appeared in known data breaches using HIBP k-anonymity lookup.

Fetches and normalizes recent SEC Form 4 insider trading records (officer/director/10%+ shareholder transactions) for any US stock ticker into structured per-transaction data.

Looks up and filters building permits across NYC, SF, Chicago, LA, Seattle, and Austin, normalizing each city's data to a common schema.

Scores a PyPI Python package for supply-chain risk based on age, downloads, install hooks, dependencies, deprecation, typosquatting distance, and provides an LLM-generated risk summary.

Searches issued US patents and published patent applications via the USPTO Open Data Portal by topic, inventor, assignee, CPC class, or filing date range, with optional plain-English AI summaries.

Returns recent earthquakes from USGS for a specified time window and minimum magnitude, including location, magnitude, depth, tsunami flag, and felt-report count.

Returns public, bank, and federal holidays for a given country and year, including date, local name, English name, holiday type, and scope.

Returns IANA timezone, UTC offset, DST status, current local time, and DST transitions for a given latitude/longitude or city name

Translates text found in images (signs, menus, screenshots, documents) using vision-OCR extraction followed by language translation, with optional layout preservation.

Converts raw HTML or a fetched web page URL into clean Markdown, stripping boilerplate and preserving structure.

Returns latest or historical FX exchange rates for 30+ currency pairs sourced from the European Central Bank, with optional amount conversion.

Normalizes Unicode text to NFC/NFD/NFKC/NFKD forms and classifies each codepoint by script, detecting homoglyphs, invisible characters, RTL markers, and BOMs

Converts a natural language description into a regex pattern with flags, explanation, and test examples, with live JS execution on sample text

Fetches an arXiv paper's abstract via the arXiv API and generates a structured LLM summary including TLDR, key findings, methods, limitations, and so-what.

Checks whether a domain name is registered and returns RDAP/WHOIS data including registrar, registration date, expiry date, days until expiry, and EPP status flags.

Decodes and inspects JWT tokens by parsing header, payload, and standard claims without signature verification

Converts a CSV file of events into an RFC 5545-compliant ICS/iCal file importable into Google Calendar, Outlook, and Apple Calendar

Summarizes up to 30,000 characters of input text into TLDR, bullet points, paragraph, or executive summary format using Mistral AI, with adjustable length, tone, and structure.

Returns recent SEC Form 4 insider transactions (buys, sells, grants, exercises) for any US-listed ticker, including officer/director/10% owner activity with share counts, prices, and EDGAR document links.

Computes the price spread, min/max prices, and spread percentage across all CoinGecko-tracked exchanges for any listed coin, identifying venues at each price extreme and the highest-volume exchange.

Analyzes a GitHub repository and returns a 0-100 health score with grade (abandoned/stale/okay/healthy/thriving) based on commit activity, contributors, license, CI, tests, and an LLM verdict.

Rewrites a long system prompt down to a target compression ratio while preserving all instructions, constraints, and examples — dropping filler, redundancy, and ceremonial language.

Queries Cloudflare 1.1.1.1 DNS-over-HTTPS to resolve any DNS record type (A, AAAA, MX, TXT, NS, CNAME, SOA, PTR, SRV, CAA, DNSKEY, DS, TLSA, HTTPS, SVCB) for a given domain, returning grouped and flat result shapes with TTLs.

Auto-detects and converts between JSON and YAML formats with configurable pretty-print and indentation, requiring no external API calls.

Generates a composite 0-100 risk score for a crypto token by aggregating honeypot detection, GoPlus Security data, and Etherscan contract age into safe/caution/high_risk verdicts with subscores.

Searches PubMed for biomedical articles by free-text query with optional date range and sort order, returning PMIDs, titles, authors, journals, abstracts, MeSH terms, and DOIs

Merges 2 to 50 PDF files (supplied as URLs) into a single combined PDF, preserving bookmarks, powered by CloudConvert.

Embeds 1–100 strings into semantic vectors using Venice-hosted models (Gemini embedding-2-preview, BGE-M3, or text-embedding-3-small), returning float vectors aligned with input order.

Returns live US airport disruption events from the FAA National Airspace System — ground stops, GDPs, AFPs, arrival/departure delays, and closures with reasons and timing

Removes the background from any image and returns a transparent PNG file.

Computes daily and annualized historical volatility (log-return standard deviation) for any CoinGecko-listed crypto token over a configurable lookback window

Scrapes a webpage and returns its title, description, metadata, headings, links, and body content as text, HTML, or markdown using Cheerio (no headless browser).

Upscales and enhances images using ESRGAN super-resolution at 1–8× scale, best for photos and illustrations, powered by fal.ai.

Captures a screenshot of any URL and returns a PNG or JPG image, with configurable viewport, retina resolution, crop/fit modes, transparent background, JPEG quality, and JavaScript wait conditions via CloudConvert.

Converts a URL, raw HTML, or webpage to a PDF with configurable page size, orientation, margins, and JavaScript rendering via CloudConvert capture-website.

Checks whether an email address is disposable, fake, or a role account by cross-referencing 100k+ known disposable domains and verifying MX records via Cloudflare DNS-over-HTTPS.

Converts PDF documents to markdown, HTML, or structured JSON with OCR-grade text extraction and structure preservation, up to 30 pages per call.

Compresses and reduces PDF file size using CloudConvert at three quality levels: ebook, printer, or prepress

Converts raw HTML or a webpage URL into clean Markdown, stripping boilerplate like navigation, scripts, and ads while preserving semantic content.

Resolves country metadata by name, ISO 3166 code, or capital city, returning capital, region, population, currencies, languages, timezones, calling code, flag emoji, and neighboring countries

Analyzes an npm package for supply-chain risk, including maintainer count, download stats, install scripts, dependency depth, deprecation, age, typosquat detection, and an LLM-generated risk summary.

Returns synonyms, antonyms, homophones, semantically similar words, or co-occurring related words for a given input word, powered by the Datamuse API.

Scrapes a webpage and returns its title, description, metadata, headings, body content, and outbound links using Cheerio (no headless browser)

Converts a datetime from one IANA timezone to another, handling DST correctly and returning formatted strings, UTC offsets, and the hour difference.

Calculates realized and unrealized profit/loss, FIFO cost basis, and current balance for any ERC-20 token in a wallet on a given EVM chain using historical Etherscan transfers priced with CoinGecko spot rates.

Fetches Wikipedia article data including title, summary, image, sections, Wikidata ID, and optional full extract for a given topic, in 300+ languages.

Fetches NIST NVD records for a given CVE ID, returning CVSS v3.1/v2 scores, severity, CWE, affected CPEs, references, exploit status, and an AI-generated exploitability summary.

Searches 30M+ Hacker News stories and comments with filtering by type, author, points, and date range, returning ranked results with metadata

Computes position changes between the two most recent 13F-HR filings for a given institutional investor CIK, returning added, exited, increased, and reduced positions with dollar deltas.

Parses a cron expression or macro and returns a plain-English explanation, frequency bucket, validation result, and runs-per-day estimate.

Extracts named entities (people, organizations, locations, dates, emails, URLs, phone numbers, monetary amounts) from text using Mistral AI, returning verbatim matches with context.

Returns assessed value, owner name, parcel ID, square footage, year built, and zoning for a property across NYC PLUTO, SF Assessor Roll, and Cook County (Chicago) datasets.

Captures a screenshot of any URL and returns it as a PNG or JPG image with configurable viewport, retina support, cropping, transparent background, and JS wait conditions.

Splits long text into chunks using fixed, sentence-aware, or paragraph-aware modes, returning each chunk's text, character offsets, and size.

Summarizes up to 30,000 characters of text into a chosen format (TLDR, bullets, paragraph, or executive summary) with adjustable length, tone, and structure, powered by Mistral.

Labels any EVM address with a human-readable name and category (CEX, DEX, NFT, MEV bot, bridge, stablecoin, etc.) and traces contract deployers via Etherscan v2.

Converts a free-form address, postal code, place name, or POI into latitude/longitude coordinates plus a structured address breakdown and bounding box, backed by OpenStreetMap Nominatim.

Resolves an IPFS CID (v0 or v1) with optional sub-path through a fallback chain of public gateways and returns the content as base64, text, or parsed JSON (max 5 MB).

Fetches, parses, and optionally summarizes SEC EDGAR filings (10-K, 10-Q, 8-K, S-1, and more) returning structured metadata, clean text, and an AI executive summary tuned to the form type.

Detects 50+ technologies used by a website (CMS, frameworks, CDN, analytics, payments, etc.) by analyzing HTTP headers, meta tags, scripts, and cookies

Converts CSV, TSV, or spreadsheet data into newline-delimited JSON (JSONL), JSON array, or column-oriented ndarray, with RFC 4180 parsing and automatic type inference

Returns NFT collection floor price, top bid, volume stats, rank, and recent sales across Ethereum, Base, Arbitrum, Optimism, and Polygon using Reservoir.

Extracts named entities (people, organizations, locations, dates, emails, URLs, phone numbers, monetary amounts) from text using Mistral LLM

Converts video files (MP4, MOV, WebM, MKV, AVI, M4V, FLV) to MP3 audio at selectable bitrates via CloudConvert, supporting files up to 500MB and 60 minutes.

Looks up WHOIS and RDAP records for a domain, returning registration date, age, expiration, registrar, contacts, nameservers, DNSSEC status, and optionally certificate-transparency data.

Returns an earnings calendar derived from SEC 8-K (Item 2.02), 10-Q, 10-K, NT 10-Q, and NT 10-K filings sourced directly from SEC EDGAR.

Generates a Conventional Commits-formatted git commit message (type, scope, subject, body, breaking_change flag) from a provided code diff

Searches FEC political contribution records by donor, committee, candidate, election cycle, state, and amount via the federal Open FEC API.

Traces every redirect hop in a URL chain, returning status codes, intermediate locations, redirect types, and response times for each hop.

Analyzes text to return overall sentiment (positive/negative/neutral) with a score from -1 to +1, plus per-emotion scores (joy, anger, sadness, fear, surprise, disgust) and optional aspect-based sentiment.

Returns upcoming and recent FDA Advisory Committee (AdComm) meeting schedules scraped from FDA.gov public pages

Fetches BLS (Bureau of Labor Statistics) public data on wages, employment levels, and growth by occupational series ID

Fetches TVL, chain breakdown, 24h/7d/30d changes, audit info, categories, and parent protocol for any DeFi protocol via DeFiLlama.

Reads the latest Base chain block, derives EIP-1559 base fee, suggests priority fees for safe-low/standard/fast tiers, and converts gas costs to USD for common transaction types.

Decodes QR codes from a public image URL and returns the embedded text strings

Converts any HTML webpage into a valid RSS 2.0 feed by auto-detecting or CSS-selector-targeted article items

Translates text into 100+ languages with auto-detection of source language, preserving Markdown, code blocks, and URLs in the output

Enumerates subdomains for a given domain by querying Certificate Transparency logs via crt.sh, returning each subdomain with first-seen, last-seen, and certificate count.

Rewrites or paraphrases text in one of 12 distinct tones (formal, casual, friendly, confident, empathetic, concise, playful, persuasive, apologetic, technical, simple, enthusiastic) with optional audience and length controls.

Looks up a US federal trademark by serial or registration number, returning mark text, owner, status, filing date, international classes, and goods/services description.

Translates text into 100+ languages with auto-detection of source language, configurable formality, and preservation of Markdown, code blocks, and URLs

Searches arXiv papers by query, author, category, or ID and returns title, authors, abstract, category, dates, PDF URL, and DOI.

Decodes a Gnosis Safe multisig transaction hash or raw calldata into its constituent fields (target, value, data, operation, gas, signers), and auto-decodes inner MultiSend batches.

Resolves ENS names to Ethereum addresses (forward) and addresses to primary ENS names (reverse), including text records like avatar, Twitter, and GitHub.

Returns part-of-speech-grouped definitions, usage examples, per-sense synonyms, IPA phonetics, and pronunciation audio URL for a given English word via the Free Dictionary API

Aggregates and returns best bridge rates across Across, Stargate, Hop, deBridge, Synapse, and CCTP for cross-chain token transfers

Rewrites AI-generated text to remove telltale LLM patterns (formulaic transitions, em-dash overuse, clichés like 'delve' and 'tapestry') across 4 tone modes

Classifies text for toxicity and harmful content across default and custom categories, returning a verdict of allow, review, or block

Transcribes audio with speaker labels using Whisper v3, returning utterances grouped by speaker with per-speaker statistics (utterance count, total seconds, word count).

Looks up CVE records from the NIST NVD, returning CVSS v3.1 and v2 scores, severity, affected CPEs, CWE class, references, exploit indicators, and an LLM-generated exploitability summary.

Detects the language of input text, returning ISO 639-1 code, full language name, confidence score, and secondary languages for mixed-language content.

Transcribes audio or video from a URL into text using OpenAI Whisper v3

Expands an image beyond its original borders using the Bria Expand AI model, generating realistic content to fill a larger canvas.

Converts latitude/longitude coordinates into a structured postal address with house number, road, city, county, state, postcode, and country using OpenStreetMap Nominatim.

Aggregates and de-duplicates keyword autocomplete suggestions from Wikipedia OpenSearch and DuckDuckGo into a ranked list for SEO research and query expansion.

Estimates the token count for a given text against a specified model's tokenizer (GPT-4, GPT-4o, Claude, Gemini) using a pure-local heuristic with ±5% accuracy versus tiktoken — no upstream API call required.

Embeds two strings using Venice's text-embedding-bge-m3 model and returns their cosine similarity as a float in [-1, 1]

Fetches recent DAO proposals, vote tallies, and voter counts for any Snapshot governance space, with an optional AI-generated TLDR summary.

Fetches structured Wikipedia article data including title, summary, image, sections, Wikidata ID, and optional full extract across 300+ languages

Reads the latest block on Base, Ethereum, Optimism, or Arbitrum to derive base fee, suggest priority fees for safe-low/standard/fast tiers, compute maxFeePerGas, and convert estimated transaction costs to USD.

Transcribes a video or podcast via Whisper and summarizes it in a chosen style (tldr, bullets, paragraph, executive, or chapters) in a single API call.

Returns registry operator, type (gTLD/ccTLD/sTLD), country, introduction year, and notes for any top-level domain

Normalizes audio loudness to a configurable LUFS target using the EBU R128 standard

Detects brand logos and visual brand elements in an image, returning confidence scores and element types (wordmark, logomark, lockup, color scheme, context)

Extracts a single frame (first, middle, or last) from a video URL and returns it as a thumbnail image.

Rewrites AI-generated text to remove LLM linguistic tells (em-dashes, 'delve', 'tapestry', formulaic transitions) across 4 selectable tones to evade GPT detectors.

Analyzes text to determine whether it was written by an AI (e.g. ChatGPT) or a human, returning a calibrated probability score, a verdict, suspicious phrases, and per-axis style signals.

Classifies a text input into one or more caller-supplied labels using zero-shot classification with calibrated confidence scores and per-label reasoning, powered by Mistral.

Converts text to speech using multiple TTS engines (Kokoro, xAI, ElevenLabs, Orpheus, MiniMax, Gemini) with 30+ voices and 6 audio formats, returning a hosted MP3 URL.

Analyzes an image using a vision LLM to produce descriptions, accessibility alt-text, OCR text extraction, keyword tags, or single-sentence captions depending on the requested mode.