Search & web

sitepeek (Tanod) robots.txt check: whether a crawler may fetch a URL under RFC 9309, with the matching rule, crawl delay and sitemaps. Read-only.

+3 more from Tanod sitepeek

Compare up to 64 total extracted spans from at most 16 labeled tools against exact code-point slices of supplied source text and declared UTF-8 hashes. Group identical source spans to expose text disagreements. No OCR/PDF execution, normalization, or completeness claim.

+1 more from Extraction consistency

Keyword ideas API: discover related keyword ideas from a seed keyword for keyword research, each with Google monthly search volume, CPC, competition, keyword difficulty and search intent. For content planning and SEO research. No ideas returns empty results.

Exa's neural web search, resold per call — no Exa account, no subscription, one USDC micro-payment. Pass query= and get ranked results: title, URL, publish date, author, relevance score. numResults=1..10 (default 5). type=auto|fast|instant. includeDomains= restricts the search. Add text=1 for matched highlights from each page (caps results at 5, priced $0.03; the 402 challenge quotes it). For agents that need what the web says today, not what a training cutoff remembers.

Pass up to 5 URLs (comma-separated) and get each page back as clean text with its title. Renders pages that build their content in the browser, which a plain server-side fetch cannot read at all — so it covers the SPAs, dashboards and app pages that url-extract deliberately refuses. Returns partial success: pages that fail come back in a `failed` list with the reason, and a call where every URL fails is not charged. For a single static page, url-extract is cheaper.

Remove all HTML tags, returning the plain text content. Entities are decoded to real characters by default (set decodeEntities=false to keep them encoded).

Extract the heading outline from Markdown — depth, text and a unique GitHub-style anchor slug for each heading. Optionally limit to a maximum heading depth.

Summarise a text's difficulty as a single consensus US grade level (averaging the five grade formulas), plus the Flesch Reading Ease, a plain-language difficulty band and an approximate age range.

Reverse lookup: resolve a Unicode character name (e.g. 'SNOWMAN', 'CJK UNIFIED IDEOGRAPH-4E2D', 'HANGUL SYLLABLE GA') to its code point. Loose whitespace/case matching. An unknown name is a successful `found:false`.

Fetch a URL and return its main article as clean HTML, Markdown and plain text.

Fetch any URL and, without paying, decode its x402 402 Payment Required challenge: price (atomic and formatted for known stablecoins), asset, network, payTo, resource, description, timeout and any input schema. Flags MPP and Bazaar discovery extensions. Returns free:true when no payment is needed. SSRF-hardened: public hosts only; private, loopback, link-local and cloud-metadata addresses are rejected.

Public business-registry record extraction (SEC / state Secretary-of-State / UK Companies House) over a buyer-supplied public records URL, wrapped in the BlindOracle trust envelope. Public records only. | BlindOracle by Craig M. Brown (craigmbrown.com) - counterparty-risk infrastructure sold to autonomous AI agents; self-built multi-agent framework (42+ agents); ERC-8004 agent passport #60979; paid via x402 in USDC on Base

+2 more from api.craigmbrown.com

Clean page extraction. Give a URL; returns the main content of the page as clean markdown (up to 8,000 characters) with navigation and ads removed. Typical response 2 to 5 seconds. If the page cannot be read you get an error and are not charged.

+3 more from bob.assimilatethis.com

Fetch a public http(s) URL and return Markdown. Pay-per-call via x402 USDC.

Query-builder helper route: autocomplete person search field values for UI/filter construction. Minimum-charge route: $0.01 on successful upstream calls.

Google Trends search interest for up to 5 terms: a relative 0-100 popularity series over time, with per-keyword average and latest value. For content timing, seasonality and demand research. Optional location (default worldwide), timeframe (7d/30d/90d/12m/5y, default 12m) and type (web/news/youtube). found=false when no data.

Fetch a URL and pull out its machine-readable structured data: all JSON-LD blocks (schema.org), Open Graph and Twitter Card tags, standard named meta tags, and inline microdata items. Also returns the page title, canonical URL, language and the distinct schema.org types found. Ideal for turning a product, article, recipe, event or organization page into structured fields without bespoke scraping.

Compute the standard English readability formulas — Flesch Reading Ease, Flesch–Kincaid grade, Gunning fog, SMOG, Coleman–Liau and ARI — along with the underlying sentence/word/syllable statistics. Empty text is a clean 400.

Live web search for agents: up to 10 results with title, URL and snippet. Use it when your agent needs to find current pages and has a wallet instead of a search API key. Charged ONLY if the search succeeds. Results pair with /v1/web/read, which returns any of them as clean markdown.

Parse an expression and return the free variables it references, without evaluating it — so an agent can discover which inputs a formula needs before supplying values. A syntax error is a clean 400.

Evaluate a JMESPath expression against a JSON value and return the selected/projected result. A malformed expression is rejected before payment.

Estimate reading and speaking time for a text at configurable words-per-minute rates (defaults: 200 wpm reading, 130 wpm speaking).

Count syllables per word and in total across the text.

Search Wikidata for entities matching a text query, returning candidate Q-ids with labels and descriptions.

Opens each URL in a cloud browser and returns one tab per URL, in input order, with the title, content type, extracted text, the accessibility tree and a link map. Results over the inline size limit return a preview per tab plus a download reference. Each call is one request.

+21 more from Retriever AI

Detect whether a User-Agent belongs to a known bot, crawler, AI agent or HTTP tool, and if so return its name and category (search, social, seo, ai, monitoring, tool, …).

Web search with clean, LLM-ready results via Tavily. Basic or advanced depth.

Plain web search. Returns ranked results (title, URL and a short snippet) for a query, up to 10, optionally limited to a publish-date range with since and until (YYYY-MM-DD). No AI-written answer, so it is fast: typically under a second. If it cannot return results you get an error and are not charged.

Pass a URL and get the page as clean text — furniture (nav, scripts, ads, footers) stripped, paragraphs preserved — plus its title, description and site name. The step every agent needs before it can reason about a page, and the one most agents can't do themselves: serverless runtimes and MCP clients have no browser and no HTML parser. Follows redirects safely, refuses non-text content, caps at 2 MB. Nothing crypto about it.

M2M memory recall: semantic search over graph_memory + farcaster_graph. Input query + fid. Output facts. Pay-per-call $0.01 USDC Base