Returns raw expansion signal rows with type, strength, polarity, source, event date and evidence URL, filtered by company, country, source type and date range. Paginated. Each call is 5 credits.

Returns people similar to a reference person, LinkedIn profile or job title, each with a similarity score from 0 to 1, narrowed by people and company filters. 25 per page. Each call is 3 credits.

Reverse lookup: returns the person matching an email, a phone, or a first and last name plus domain or company, with their Pubrio id, base details and match metadata. 1 credit per match; no match costs nothing.

Run before buying, accepting or recommending a token on Base. Answers SAFE, CAUTION or AVOID with the exact reasons, from a simulated buy and sell, the contract's verified code and market data.

Run before sending ETH or USDC or paying gas, to confirm the wallet really has enough. Never trust a balance remembered from earlier in a chat. Reads a wallet's ETH and native USDC balance on Base directly from the chain, at one block.

Returns one company's expansion profile by domain, LinkedIn URL or Pubrio id: stage per market, signals, market presence, stage history and expansion score, for one market or all markets, with an optional AI summary against your query. Each call is 5 credits.

Bulk reverse lookup of up to 50 people in one synchronous call, results in input order. Unmatched and duplicate inputs are dropped and not charged. 1 credit per match.

Returns ready-to-render expansion summaries for a list of companies, by domain, LinkedIn URL or Pubrio id, for side-by-side display. Each call is 5 credits.

Citation check: takes a claim and a cited source (URL or DOI) and returns supported, contradicted or not_addressed with the verbatim passage as evidence. The source is fetched with a full browser render. Dead, paywalled or PDF-only sources return source_unavailable, never a false contradicted.

Textus: structured content with editable baseline direction, characters and scenes for audio, games and media production. One purchase returns one selected text unit as JSON; no audio or provider execution. Find IDs, metadata, prices and availability: https://api.textus.dev/api/v1/resources.json

Generate images from a text prompt with Google's Nano Banana 2 (Gemini 3.1 Flash Image) — fast, high-quality generation. Returns hosted image URLs.

Generate images from a text prompt with Google's Nano Banana 2 (Gemini 3.1 Flash Image) — fast, high-quality generation. Returns hosted image URLs.

Research bundle: takes a query and returns up to 10 deduplicated, rendered web sources as one token-budgeted Markdown bundle (about 8,000 tokens), one source per domain. Combines web search with rendered-page fetching in a single call. Needs at least 3 fetched sources; the response reports sources_requested and sources_returned. Optional region=tw selects a Taiwan, Traditional Chinese lane.

Returns the best bid and ask with sizes and the observation timestamp for one stock. 404 when the symbol is unknown, 503 when no serviceable quote exists.

Google organic ranking check: finds whether and where a domain ranks for a keyword, using DataForSEO SERP data. Returns found, position, URL, title and total organic results.

AI answer visibility check: sends a real query to a live AI model with web search enabled and checks the response for a mention of the business and a citation linking to its domain. Returns mentioned, cited, citations and an excerpt.

Taiwan government reference data: ARC and visa rules and all 15 WDA work-permit categories, translated to English and structured into labelled sections. With no parameters it returns an index. Republished under Taiwan's Government Open Data License with attribution. Machine-translated; the linked original is authoritative.

Returns a full-page PNG screenshot of a URL, captured in a real rendered browser (JavaScript executed). The response body is the image, not JSON. Cached for 24 hours per URL; X-Cached and X-Fetched-At response headers report cache status and fetch time.

Returns a PDF of a full web page, rendered in a real browser (JavaScript executed). The response body is the file, not JSON. Cached for 24 hours per URL; X-Cached and X-Fetched-At response headers report cache status and fetch time.

Extracts structured JSON from text already in hand, matching a JSON Schema supplied by the caller, using AI. No browser render, so it costs less than /extract-url. Takes content and schema; returns data and the model used.