Web & Data

Recent news headlines + heuristic bull/bear sentiment for a ticker or topic (Google News, no LLM).

Newest headlines from a deduped multi-source pile (GDELT + Hacker News + major RSS + Google News), newest first — filter by category, source, and language

Read a public EVM wallet balance on Base or Ethereum and determine whether its Base USDC balance can cover a suggested $0.001 x402 payment. Returns a coherent provider snapshot, USDC runway, shortfall, native-gas presence, and a handoff to the full funding-readiness doctor without signing, broadcasting, or handling private keys. Balance-only readiness does not prove client, signing, spend-cap, or retry capability. GET variant for agents, crawlers, and wallets that prefer query parameters over...

List & search Pokemon species by name, type or Pokedex range — paginated structured JSON reference for game agents

Looks up classic and contemporary books by title, author, or ISBN via Open Library (748M+ editions). Returns publication year, subjects/genres, page count, cover images, and links to read online via Project Gutenberg or Internet Archive. Useful for research agents, reading recommendation workflows, literature analysis, and bibliographic data enrichment.

Extracts structured breadcrumb navigation from a URL. Returns domain, ordered path segments with human-readable labels, query parameters as key-value pairs, and a formatted breadcrumb trail string. Identifies numeric IDs vs. named path segments. Pure URL parsing — zero external calls. Useful for agents that process sitemaps, navigation menus, or need to understand page hierarchy.

Extract structured company intelligence from any public website. Returns company name, description, logo, emails, phones, address, founded date, social links (Twitter/LinkedIn/GitHub/etc.), and raw OpenGraph + schema.org/Organization data. Pure HTML extraction — no external APIs. $0.003 hedge against orbisapi web-scrape-company at $0.005.

Hacker News story and comment search via Algolia. Returns titles, scores, comment counts, authors, and URLs for posts matching the query. Filter by type (story/comment) and date range (day/week/month/year/all). Sorted by relevance by default; use sort=date for newest-first. Useful for tech news, community sentiment, or discovering discussion threads about a topic.

Transcribe audio from any publicly accessible URL using OpenAI Whisper. Supports mp3, mp4, m4a, wav, webm, ogg, flac, and wma up to 24 MB. Returns the full transcript text, detected language, and estimated duration in seconds. Optionally accepts an ISO 639-1 language hint to improve accuracy. Useful for processing voice memos, meeting recordings, podcast snippets, interview clips, and audio attached to social media. Undercuts orbisapi.com audio-transcription-api by 24%.

Search the Federal Register for Rules, Proposed Rules, and Notices matching a keyword — returns title, abstract, document type, agencies, significance, effective date, and links. No API key. $0.002/call.

Academic paper search across 250M+ works via OpenAlex (free, no key). Returns top papers with title, authors, year, DOI, citation count, open-access status, and primary research topic. Covers all disciplines: AI/ML, medicine, physics, economics, law, biology, and more. Supports relevance, citation-count, and recency sorting; open-access filtering; and year-range constraints. Use for literature review, prior-art search, citation building, or finding the seminal papers in any field.

AI-powered URL content analysis. Fetches a URL, extracts the readable article text, and returns structured intelligence: 2–3 sentence summary, key points, named entities with types, sentiment score, topic tags, content type classification, and credibility signals (has author/date/sources). Use for content intelligence pipelines, research synthesis, or automated brief generation. $0.012/call.

Extracts structured content from any public URL: page title, meta description, H1-H3 headings, all links (with text and internal/external flag), and a 500-character text preview. Useful for research agents following link chains from on-chain data, auditing page structure, or seeding downstream text-generation calls.

Parse and explain any Unix cron expression in plain English. Returns human-readable schedule description, field breakdown, and the next N run times (UTC). Supports @yearly/@monthly/@weekly/@daily/@hourly shortcuts. Zero external calls.

Wikipedia article lookup and search. Given a search query, returns the top matching Wikipedia articles with title, plain-text extract (~800 chars), description, thumbnail URL, page URL, and last-modified date. Use for rapid factual lookup, entity enrichment, concept explanation, or pre-flight research on any topic. No API key required.

Extracts all hyperlinks from any public webpage. Returns href URLs normalized to absolute URLs with visible link text. Filters out javascript:, mailto:, data: schemes. Optionally restrict to same-domain links, deduplicate, or include #anchor links. Useful for crawlers, sitemap builders, link graph analysis, and content audits.

Fetches and parses any public RSS 2.0 or Atom 1.0 feed. Returns feed metadata (title, description, language, last updated) and structured items (title, link, 400-char summary, author, published date, GUID). Useful for monitoring news, blog posts, GitHub release notes, Reddit RSS, arXiv category feeds, or security advisories.

Country information lookup by name, ISO code (alpha-2 or alpha-3), or capital city. Returns: official name, ISO codes, capital, population, area (km²), region, languages, currencies (code + symbol), borders, calling code, timezones, and flag image URL. Useful for international business agents, geographic enrichment, currency identification, and compliance workflows.

Extracts all hyperlinks from a webpage. Fetches the target URL, resolves relative links to absolute URLs, and classifies each as internal (same domain) or external. Filter by all/external/internal, cap results with limit. Returns page title, total link count before filtering, and a structured array of {href, text, is_external, domain}. Priced at $0.004 — 20% below orbisapi web-scrape-links ($0.005/call). Upstream: direct HTTP fetch, no API key.

Fetches any public URL and returns the full readable article text as clean Markdown, stripped of navigation, ads, and boilerplate. Returns title, published date (if available), and the complete body ready for LLM summarization, analysis, or RAG ingestion. A $0.004 alternative to exa.ai/contents ($0.007) and web-read ($0.016) for the same full-text extraction primitive.

Returns current US corporate credit spreads from ICE BofA indices via FRED (free, no API key): High Yield OAS, Investment Grade OAS, and BBB (lowest IG tier) OAS. Includes HY-IG differential and risk regime classification. Pairs with treasury-yields for complete fixed-income discount rate construction.

Extracts and parses JSON from mixed-content text. Handles LLM output with JSON embedded in prose, code fences (```json), trailing commas, single-quoted strings, JS-style comments, and bare object keys (JSON5-style). Returns the parsed data, a cleaned JSON string, extraction method used, and any repair applied. Pure text processing — zero external API calls.

Search HuggingFace Hub for ML models. Specify a keyword (e.g. 'bert', 'llama', 'stable diffusion') and optional task filter (e.g. 'text-classification', 'text-generation', 'image-classification'). Returns top results sorted by downloads or likes, including model ID, author, pipeline task, framework library, download count, likes count, and tags.

Searches 5M+ US court opinions (SCOTUS, federal circuits, district courts, state courts) via CourtListener. Returns case name, court, date, citation, docket number, and a text snippet. Filter by court level, date range, or judge name. Useful for legal research agents, contract risk analysis, precedent lookup, and regulatory compliance workflows.