Explore APIs

x402 seller due diligence: everything in x402-preflight PLUS on-chain proof of real demand (USDC actually received by the seller's payTo in the last 24h: payments, distinct payers, volume) and a Bazaar listing-truth audit (is this URL listed, does the listed price match what it charges). Reveals dead, fake or mispriced sellers before payment.

Phone number intelligence for agents: pass any number and get back whether it is a valid E.164 number, its country, the carrier that holds it, and the line type — mobile, landline, or VoIP. One paid call, no account, no API key. Decide whether a number is worth calling or texting, or flag a VoIP number before you trust it. Free-format numbers are normalized; a syntactically bad number is rejected

Keyless web search for agents: pass a query, get a concise web-grounded answer plus the top source results (title, URL, snippet). Live web access, no API key or account needed. JSON response. $0.008 USDC per call, pay via x402, no API key.

Look up a product barcode number: identify and validate EAN-13, EAN-8, UPC-A, UPC-E, GTIN-14, ISBN-13/10, ISSN and ISMN codes (wrong check digits are reported with the correct digit), convert between forms (GTIN-14, EAN-13, UPC-A, ISBN-10 <-> 13) and decode the GS1 prefix. Then fetch product data from open databases: name, brand, quantity, categories, image, labels, Nutri-Score (food, cosmetics, p

EDINET filing calendar: annual securities reports (yukashoken hokokusho) filed by Japanese listed companies in a date range. Returns company code, EDINET document ID, fiscal period, accounting standard (JGAAP/IFRS/USGAAP) and the exact filing timestamp for each report — track new disclosures, trigger data pipelines, or find which Japanese stocks just reported. Query: /v2/calendar?from=2026-06-20&t

Amazon price + price history + monthly sales by ASIN. One call, one ASIN: current buy box, Amazon, lowest new and used prices PLUS 30/90/180-day price averages and the typical 180-day price range, estimated monthly sales, sales rank, rating, review count, offer count, brand, category, image. From tracked marketplace history, not a page scrape. 11 Amazon marketplaces (domain=us, uk, de, fr, jp, ca,

Resize an image at a URL to a width, a height or both, keeping the aspect ratio inside the box by default (cover, contain, inside, outside or fill), optionally converting it. Returns a signed URL of the result, valid for an hour.

Encode text or a URL as a QR code: PNG (base64 and a data URL) or SVG, with the size and error correction level you choose.

Wikipedia summary API - the plain-language summary of the best-matching Wikipedia article for a topic or query, in any language. Returns the article title, short description, summary extract, thumbnail image and canonical URL. A wikipedia / encyclopedia lookup for agents grounding answers on a topic. No match returns found:false. Data source: Wikipedia (CC BY-SA 4.0).

Fetch a public URL and return its main readable content as Markdown or plain text, plus title, author, published date and site name. Strips navigation, ads and boilerplate. No JavaScript rendering by default (cheaper than readfetch_scrape); one Chromium fallback when plain HTTP is blocked. Prefer readfetch_fetch when you need raw status, headers, or non-HTML bytes; readfetch_scrape when the articl

Detects words and lines of text in a photo or screenshot (JPEG or PNG at a URL) with Amazon Rekognition, up to 100 words. Suited to signs, product labels and UI text; for documents use /v1/ocr.

YouTube video transcript extraction with per-segment timing. Returns the caption track as timed segments (start, duration, text) plus the full plain text, or as SRT subtitles. Uses the creator's manual captions when they exist and falls back to YouTube automatic captions. Never downloads media, so it stays fast and light. Cached per video and language. Use it to summarize a talk, quote a video wit

utilpeek (Tanod) file upload: PDF, Office, image, audio, CSV, JSON or text stored for up to 24 hours and returned as a download link. Use when an agent holds a file and must give the next tool a URL for it: a link that expires within 24 hours.

Agent tools search: One search across ~13,000 agent tools: 4,259 MCP servers (awesome-mcp-servers, MIT) and ~8,900 x402 pay-per-call APIs (nohumans.directory, CC BY 4.0). Filters: q, type=mcp|x402, limit<=100.

Read any public web page as clean Markdown: main-content extraction (navigation, footers, scripts and hidden elements removed), headings, lists, tables, code blocks, links (absolute, deduplicated), images, plus title, description, canonical URL, language, author and publish date, and a content hash + fetch time as evidence. JSON and plain-text URLs are returned as-is. No JavaScript rendering. $0.0

Web search for AI agents: search the web for any query and get ranked results (title, URL, snippet) plus the full text of the top 1-3 pages, each read in a real browser, as clean JSON. Filter by date, site or news. Find current, up-to-date information online and answer with sources in one call, instead of a search followed by separate page reads.

Maps a seed and asset category to declared silhouette, material, palette, symmetry, and detail parameters. The result is a machine-readable design specification only; it does not render geometry, textures, animation, or production-ready art.

Twitter/X Posts & Tweets Search: the latest public tweets on X matching a keyword or phrase, newest first. Per tweet the text, author handle and name, verification, followers, likes, replies, retweets, views, language, time and URL in one schema shared by every platform; parent_id names the replied-to or quoted tweet; next_cursor gives the next page. For social listening, brand monitoring and rese

Wikipedia article summary — plain-text and HTML extract, short description, thumbnail and Wikidata id for any title, in a clean normalized envelope with CC-BY-SA attribution.

Use when an agent or user wants a clean original joke — family-friendly humor from the Synthetic Universe. Pure compute, no upstream. Optional session id avoids repeats. Use permitted: build it into your products and services, paid or free. Raw redistribution of the feed itself is prohibited.

Munition Firecrawl Scrape: x402-paid web page scraping for AI agents. Pay 0.015 USDC per URL to get the page rendered with JavaScript and converted to clean markdown (optionally raw HTML, links, summary). Use when a plain fetch returns empty, JS-only or 403 content.

Currency exchange rates and conversion for about 30 currencies (USD, EUR, GBP, INR, JPY, CNY ...) from the European Central Bank reference rates of the latest working day.

Generate sentence embeddings (all-MiniLM-L6-v2, 384-dim, mean-pooled + L2-normalized) for up to 96 texts per call. Drop-in compatible with any all-MiniLM-L6-v2 index. Pairs with /chunk for a complete RAG pipeline.

Requires shares totaling exactly 10,000 basis points, floors each proportional minor-unit allocation, and distributes leftover units by descending fractional remainder with a stable name tie-break. It performs arithmetic only and does not transfer funds, interpret contracts, withhold tax or resolve rights ownership.

The ranked coin feed pump.fun's home page loads: each coin's mint, name, image, market cap, protocol (pump bonding curve or AMM), creator, creation time and activity figures. A broader, slower-moving list than the trending board. Use for market-wide scans.

Recall tenant memory

Generate images from a text prompt with xAI's Grok Imagine quality tier. Returns hosted image URLs.

Start the "Instagram Scraper" Apify actor (apify/instagram-scraper). Scrape and download Instagram posts, profiles, places, hashtags, photos, and comments. Get data from Instagram using one or more Instagram URLs or search queri…

Extract text and metadata from PDF documents. Supports page-specific extraction and returns structured content.

One 0-100 safety score for a Base token before an agent buys: liquidity, top-10 holder concentration, owner powers in the bytecode, upgradeability (EIP-1967), explorer/Sourcify verification, creator history, small-trade price impact. Verdicts: safe/caution/danger per component, unknown when a check could not run, and not_a_token with no score when the address deploys no code.