Web Search Research x402 offerings

For JavaScript-rendered pages & pages blocking native Minifetch that obeys robots.txt: fetch thru external rotating proxy, render JS in headless browser & extract same live, clean markdown as /extract/url-content, after nav, ads, and scripts are stripped. Optional ?country= (2-letter code) to fetch from a specific country. Not cached. Does not obey robots.txt. Supports GET & POST

For pages blocking native Minifetch that obeys robots.txt: fetch thru external rotating proxy & extract same categorized links as /extract/url-links (internal, external, anchor) with rel attributes, anchor text and summary stats. No JavaScript rendering; use /proxy/render for that. Optional ?country= (2-letter code) to fetch from a specific country. Not cached. Does not obey robots.txt. Supports GET & POST

For pages blocking native Minifetch that obeys robots.txt: fetch thru external rotating proxy & extract same live, rich structured metadata as /extract/url-metadata (headers, favicons, title, description, canonical, meta tags, OG, JSON-LD, etc). No JS render; use /proxy/render for that. Optional ?country= (2-letter code). Same ?fields= and ?omitEmpty= params. Not cached. Supports GET & POST

For JavaScript-rendered pages & pages blocking native Minifetch that obeys robots.txt: fetch thru external rotating proxy, render JS in headless browser & extract same live, rich structured metadata as /extract/url-metadata (headers, favicons, title, description, canonical, meta tags, OG, JSON-LD, etc). Optional ?country= (2-letter code). Same ?fields= and ?omitEmpty= params. Not cached. Supports GET & POST

For pages blocking native Minifetch that obeys robots.txt: fetch thru external rotating proxy & extract same share-preview fields as /extract/url-preview (title, description, Open Graph, Twitter Card, favicons). No JavaScript rendering; use /proxy/render for that. Optional ?country= (2-letter code) to fetch from a specific country. Not cached. Does not obey robots.txt. Supports GET & POST

For JavaScript-rendered pages & pages blocking native Minifetch that obeys robots.txt: fetch thru external rotating proxy, render JS in headless browser & extract same categorized links as /extract/url-links (internal, external, anchor) with rel attributes, anchor text and summary stats. Optional ?country= (2-letter code) to fetch from a specific country. Not cached. Does not obey robots.txt. Supports GET & POST

For link audits: Extracts and categorizes all links on a page (internal, external, anchor) for SEO audits and site-structure analysis. Returns nofollow/rel attributes, anchor text, image links missing alt text, and external domain distribution. Obeys robots.txt. Supports GET & POST

Extracts live, clean markdown from a page after nav, ads, and scripts are stripped. Can also be used by site owners auditing AI readability: see what survives for AEO, GEO and AI search. Obeys robots.txt. Supports GET & POST

For pages blocking native Minifetch that obeys robots.txt: fetch thru external rotating proxy & extract same live, clean markdown as /extract/url-content, after nav, ads, and scripts are stripped. No JavaScript rendering; use /proxy/render for that. Optional ?country= (2-letter code) to fetch from a specific country. Not cached. Does not obey robots.txt. Supports GET & POST

Keyword web search by Ceramic.ai returns ranked results as structured JSON: a title, URL, & text snippet per result. The discovery entry point for agents & RAG pipelines: search to find relevant pages, then go deeper on any result with the paid /extract endpoints. Pay-per-search. Use ?limit= to cap results (1-10) and ?descriptionLength= to trim each snippet (0 = titles & URLs only) for desired token efficiency. Supports GET & POST

Use when the user needs to see a page as it actually renders — visual proof, a preview, a layout check, or a record of what a page showed. Takes a screenshot with a headless browser (rendering JavaScript). JSON body { url } (public http/https); optional full_page (default false), width (320–2000, default 1280), height (240–2000, default 800), format png|jpeg (default png). Returns image_base64. Private/internal targets are blocked.

Use when a page will not read correctly with a plain fetch — a single-page app, content loaded by JavaScript, or markup so noisy the text is buried. Fetches it with a headless browser (rendering JavaScript) and returns the main readable content as clean Markdown — ideal to feed an LLM as context. JSON body { url } (public http/https), optional include_images (default false). Returns title, final_url (after redirects), markdown and word_count. Private/internal targets are blocked.

MSHA mining fatalities feed, newest first, including OPEN preliminary notices (scraped 32–180 days ahead of the official dataset and flagged as preliminary). Filter by date, state, commodity.

Claude pricing table: what every Claude model costs per token right now — Claude 5, Claude Opus, Claude Sonnet, Claude Haiku — from Anthropic and every provider that resells or hosts them (Bedrock, Azure, Vertex, OpenRouter, Fireworks). Input, output, cache and batch USD per 1M tokens, sorted cheapest first, so you can compare Claude inference cost across hosts and pick the cheapest place to run one. Normalized, cross-checked, refreshed hourly.

Largest-context-window leaderboard: the 100 AI models with the biggest context windows, ranked descending — how many tokens each model can actually take, its max output tokens, modality support, and what those tokens cost. Compare GPT-5, Claude 5, Gemini 3 Pro, Llama 4 Scout and 2,000+ more on context capacity and price together. Refreshed hourly.

DeepSeek pricing table: what every DeepSeek model costs per token right now — DeepSeek V4, DeepSeek R1, DeepSeek Coder — from DeepSeek and every provider that resells or hosts them (Bedrock, Azure, Vertex, OpenRouter, Fireworks). Input, output, cache and batch USD per 1M tokens, sorted cheapest first, so you can compare DeepSeek inference cost across hosts and pick the cheapest place to run one. Normalized, cross-checked, refreshed hourly.

Mistral pricing table: what every Mistral model costs per token right now — Mistral Large, Mixtral, Codestral, Magistral — from Mistral AI and every provider that resells or hosts them (Bedrock, Azure, Vertex, OpenRouter, Fireworks). Input, output, cache and batch USD per 1M tokens, sorted cheapest first, so you can compare Mistral inference cost across hosts and pick the cheapest place to run one. Normalized, cross-checked, refreshed hourly.

Cheapest long-context LLM leaderboard: the 50 lowest-cost AI models with a 200,000-token context window or larger, ranked by inference cost per token — Claude 5, Gemini 3, GPT-5, Llama 4 Scout and more across 70+ providers. Input, output and cache USD per 1M tokens with exact context window and max output tokens joined in. Refreshed hourly.

Fetches the Monexus Terminal article index with filters for category, limit, and date.

Call when you need the text of one public http(s) page. Required query url. SSRF-gated. Not a search engine and not a GitHub clone.

Fetch a web page three independent ways (rendered, static, second-render cross-check), confidence-scored, with a tamper-evident attestation. Refuses protected, login and CAPTCHA pages without charging. Use when your agent needs web content it cannot fetch itself — JavaScript-rendered pages, static pages, or paginated listings — and needs to show its principal the evidence. Every extraction runs three independent fetches (rendered, static, second-render cross-check), is confidence-scored, and is written to a public tamper-evident attestation chain. Protected, login, and CAPTCHA pages are explicitly refused and never charged. Money Mind sells judges, not strategies. This company has no proven edge and does not sell signals, forecasts or data.

Verify a claim against 2-4 source URLs: cited passages, cross-source agreement, verdict: consistent | conflicting | insufficient_evidence. Evidence, not truth. Use when your agent needs to verify a claim against independent sources before acting on it. Give a claim plus 2-4 source URLs; each source is extracted (three-fetch cross-check, confidence-scored), the checkable figures are compared across sources, and you get the cited passages plus a verdict: consistent (sources report the claim's figures), conflicting (sources agree with each other on different figures), or insufficient_evidence. This sells evidence, not truth — the verdict describes source agreement and says so. One $0.03 call replaces three separate extractions plus your own comparison. Money Mind sells judges, not strategies. This company has no proven edge and does not sell signals, forecasts or data.

Verify a claim against 2-4 source URLs: cited passages, cross-source agreement, verdict: consistent | conflicting | insufficient_evidence. Evidence, not truth. Use when your agent needs to verify a c…

Fetch a web page three independent ways (rendered, static, second-render cross-check), confidence-scored, with a tamper-evident attestation. Refuses protected, login and CAPTCHA pages without chargin…