Web Search Research x402 offerings

Cross-encoder reranking of retrieved documents by relevance to a query, at fixed model precision. POST /v1/reranking with {query, documents} and get the reordered ranking as one JSON object. Deterministic. Pay per request in USDC; no account, no API key.

Fetch a URL and get it back ready for an LLM or RAG pipeline: clean markdown (headings and links kept), plain text, the title, the page's links and fetch metadata. POST /v1/extract with {url} and optional formats and max_chars. JSON in, JSON out, errors included. Pay per request in USDC; no account, no API key.

Fetch a web page and return structured data: standard fields (title, description, language, h1, OpenGraph, author, published), headings, JSON-LD and a text excerpt. Optional schema picks named fields or applies regexes to the HTML. POST /v1/url with {url}. JavaScript is not executed; pages are cached for 10 minutes. Pay per request in USDC; no account, no API key.

Screen for SAM.gov exclusions: excluded parties by name, UEI, or CAGE with termination dates. POST /v1/debarment with a {name} and get matches and the lists checked as one JSON object. Deterministic per daily cache. Pay per request in USDC; no account, no API key.

Load a URL in headless Chromium with JavaScript executed and get the rendered page back for an LLM or RAG pipeline: markdown, text, HTML, title, links and an optional screenshot. POST /v1/render with {url}. Robots.txt is honored and private addresses are refused. JSON in, JSON out, errors included. Pay per request in USDC; no account, no API key.

Fetch a URL and return its title and readable body text, with navigation and boilerplate stripped. POST /v1/url with {url} and get one JSON object back. The text is flattened: headings, links and lists are not kept as markdown structure. Responses may be served from cache. Pay per request in USDC; no account, no API key.

Web search built for agents, served over Pocket Network: POST /v1/search with a query and an optional max_results, and get the results back as one JSON object. Results vary run to run, as web search does. GET /v1/version, /v1/health and /v1/capabilities describe the service. Pay per request in USDC; no account, no API key.

Fetch a web page and return its title and raw HTML (up to 200,000 characters). No browser engine is used and JavaScript is not executed, despite the name. POST /v1/render with {url} and get one JSON object back. Pages are cached for 5 minutes. Pay per request in USDC; no account, no API key.

Summarizes podcast transcripts into concise, readable takeaways.

Convert a Twitter/X, LinkedIn, Medium, Substack, Beehiiv, or public Notion page URL into clean markdown.

What changed since your last read: one page of PreReason's change events after a cursor. A briefing's state label, a tracked release, and, for the companies you name (?companies=, at most 200), newer figures, a situation turning on or filed news (?kinds= narrows it). Each event says what changed, when it happened, when PreReason knew it and where to fetch it, never a value. With no cursor it answers the last 24 hours; each answer's next_cursor is where the next read starts.

Extracts key quotes from press releases with AI.

Use this when you need a cited research brief on any topic: searches the live web, reads 3-6 sources, and returns a structured JSON brief — summary, key findings each linked to its source URL, confid…

Use this when you need a fast cited answer, not deep research: one live web search, relevance-ranked sources, snippet-level digest with 3-6 cited key points in seconds. No API key, no signup — x402 p…

Bulk price on page: up to 20 urls in one call, processed concurrently, results returned in input order with a per-item error field and a count of failures. Same answer per item as the single /price-on-page endpoint (Price on a page). Batch enrichment for agents that hold a list. $0.10 per batch.

Change watch for price on page: runs the check and compares it with the result from your previous call for the same url (kept 90 days), returning the current answer, changed true/false, exactly which fields changed with before/after values, and when it was last seen. Call it on a schedule to be told when price on a page changes. $0.02 per check.

Korean semiconductor industry news translated to English. Covers Samsung Electronics, SK Hynix, HBM/DRAM/NAND memory products, foundry, and supplier ecosystem. Headlines from Korean tech press (전자신문/디지털타임스/한국경제) auto-classified and translated by Claude. 5-min cache.

Korean K-pop news translated to English. Naver-aggregated headlines from Korean media (Yonhap, Korea Economic Daily, Chosun Ilbo, etc.) auto-classified for K-pop relevance and translated by Claude. Returns title_en + summary_en + source_en + original Korean references. 5-min cache. limit=1..10.

Korean K-pop news with AI synthesis. Returns the headline list (English-translated) PLUS a Sonnet 4.6 analysis: overall_sentiment, key_themes, trending_entities, and a paragraph summary. For agents tracking Korean entertainment industry pulse. 5-min cache.

Korean semiconductor industry news with AI market synthesis. Returns headlines (English-translated) PLUS Sonnet 4.6 analysis: overall_sentiment, key_themes, trending_entities, market_signal (bullish/bearish/neutral), and a paragraph synthesis. For agents tracking Korean memory/foundry industry signals. 5-min cache.

Content fetch API. Pay per request via x402 and receive the article text together with its Proofivy content-id (CID), a cryptographic commitment to the exact bytes returned.

Content fetch API. Pay per request via x402 and receive the article text together with its Proofivy content-id (CID), a cryptographic commitment to the exact bytes returned.

Extract bounded title, description, canonical, Open Graph, JSON-LD, and link metadata from a public URL.

Fetch a public HTTP(S) page with private-network blocking and return bounded Markdown-like content plus title, description, Open Graph image, status, and URL metadata.