Web & Data

Extract structured metadata from a page: title, description, OpenGraph, JSON-LD, headings, links

VAPE rug_pull_alert — Owner-power / rug-risk flags (mint, blacklist, pausable transfers, LP concentration). Real on-chain/market data, no simulation.

Validate a ULID and extract its embedded timestamp and time/random components.

Keyword search over the retained multi-day news archive (scanned span reported in depth + freshness.coverage) — AND-ed terms plus quoted phrases over title + summary + domain + categories, source/lang/date filters, recency or relevance sort, up to 100 per call

Volume of matching headlines over time (hour/day buckets) plus average sentiment per bucket with its provenance (GDELT document tone vs local lexicon, never a silent null) — a local sentiment timeline over the retained multi-day archive

DNS lookup — Resolve DNS records via DNS-over-HTTPS.

robots.txt / llms.txt — Fetch and parse a site's robots.txt and llms.txt.

Current headline values (price, demand, and where published carbon intensity) for all 25 GridHub zones in one call. Use to compare or rank grids right now; rebuilt every 5 minutes. For one zone with context use /v1/brief/{zone}.

Generate one or more lexicographically sortable ULIDs (monotonic within a call).

Creates a fresh browser context for one HTTPS page, obeys robots.txt, refuses login walls, CAPTCHAs and paywalls, renders JavaScript, returns a sanitized DOM and/or base64 PNG, and destroys the context after the request. There is no reusable cookie jar, session handle, login automation or navigation step list. The runtime ceiling is 60 seconds and the returned DOM is capped at 1 MiB.

Draws a random integer within any min-max range using HMAC-SHA256 over a client-supplied seed, revealing the server seed so the draw can be independently recomputed and verified afterward. For raffles, matchmaking, tie-breaking between competing agents, and any process where 'trust me, it was random' isn't good enough.

Free-text web search to ranked organic results, sponsored rows excluded. Returns results[] with rank, title, real destination url, displayUrl and snippet, plus count and attribution. Requires ?q= (max 500 chars); optional ?count=1..25 (default 10). Errors: 400 missing_query|query_too_long|bad_count, 404 no_match, 503 upstream_unavailable. Results are NOT fetched or verified - to read one, pass its url to /markdown. For what people type rather than what ranks, use /suggest.

Turn a 4-byte function selector or full calldata into its human-readable function signature, and (when calldata is present) the decoded argument values. Supply an `abi` for a fully deterministic offline decode, or omit it to look the selector up in public signature registries. Reads with automatic fallback.

Retrieve bounded public web content with source, timestamp and cryptographic hashes

Proposer feed for prediction-market settlement: SEC EDGAR official facts mapped to a machine-friendly answer / timestamp / evidence shape for automated proposal and dispute flows. Three separated times, explicit resolution_status, candidate sources listed. We describe what was filed - the adjudication stays yours. Descriptive only, never a forecast or advice. No data -> 422 no_charge, not billed. $0.01 per call.

Compute the level monthly payment, total paid and total interest of a fully-amortizing loan. Handles a 0% rate. Set schedule=true to also return the full month-by-month amortization table.

Search the constants table by key, symbol or name substring. Exact matches rank first, then prefix matches, then substring matches. Returns up to `limit` constants (default 20) plus the total match count.

Sniff which structured format an unknown document is. Because formats overlap, returns every plausible format with a 0–1 score plus the best guess — honest about ambiguity rather than pretending certainty.

Reddit subreddit feed scraper — the latest posts from any subreddit (r/<name>) for AI agents, sorted by hot, new, top, rising, or controversial with an optional time window. Returns each post’s title, author, score, comment count, permalink, external link + domain, flair, NSFW flag, and timestamp. No login, no Reddit API key. Monitor communities, track trending posts, and feed social signals to agents.

Fetch a public article or PDF and return clean Markdown plus title, byline, siteName, excerpt and wordCount. HTML is extracted with Firefox reader-mode rules; PDFs return their text layer. Requires ?url=<public http(s) URL>. Errors: 400 missing_url|bad_url, 403 blocked_private (private and internal hosts refused, every redirect hop re-checked), 413 too_large above 8 MB, 422 no_text_layer for scanned PDFs or not_extractable for app shells, 504 fetch_timeout. To FIND urls, use /search.

Looks up one IPv4 or IPv6 address against the two authoritative, keyless public sources: RDAP registry data (who holds the block, in which country, since when, abuse contact) via the ARIN-to-RIR referral chain, and the origin ASN from Team Cymru's DNS service. No city-level geolocation and no VPN/proxy detection — those need a licensed commercial database this endpoint deliberately does not use. Built for going through a list of IPs one at a time.

One curious but true fact, in English or Spanish, filtered by topic: science, computing, history or nature. Takes parameters, returns an id so you can avoid repeats across a session, and accepts a seed for a reproducible answer.

Apply a Unicode normalisation form (NFC, NFD, NFKC or NFKD) to a string. Reports whether the text changed and its length in UTF-16 code units and code points.

Fuzzy-search countries, currencies and languages by name or code. Scope with `registry` (defaults to all). Matches are ranked exact-code/name → prefix → substring.