Web & Data

Readable text extraction: fetches one public HTML page server-side and returns clean plain text with the page title, final URL, HTTP status, content type, SHA-256 of the response body, a timestamp and a tamper-evident receipt. Single GET query parameter (url). Text is capped at 12000 characters; no JavaScript is executed.

Generate a random handshake sequence (numbers and symbols) for secure connections.

Generate a summary of a given text using the GLM chat model.

Execute Python code in a real, isolated E2B cloud sandbox (not a subprocess) and return stdout/stderr/results.

Fetch a file's contents from a GitHub repository at a given ref.

Scrape a URL and return clean Markdown (via Firecrawl). Priced above Firecrawl's own per-scrape cost.

Search current Google News results. Returns headline, source/domain, URL, snippet, publication time, image and rank. For current-events, monitoring and research agents.

Web search over Google's organic results (SERP API). Returns ranked organic results with title, URL, domain, snippet and breadcrumb for any query. For research, retrieval-augmented agents, web lookup, fact-finding and general search.

Newest-first global feed of FOMO theses (short trade write-ups: token, author, PnL, holdings). Use only when you have no token or author filter. Page with ?before= from nextBefore; stop when hasMore is false. Dedupe items on id. Prefer the token/user-scoped routes when you know those.

IBAN validation (ISO 13616 + mod-97 checksum, 70+ countries). POST {iban}. Returns {valid, country, bban?, reason?}.

Tight factual summary of raw text/HTML. POST {text, focus?}. Returns {summary, key_facts[]}.

Structured JSON extraction from messy text/HTML. POST {text, fields[]}. Returns {data}.

Sanctions screening: check any person or company name against official OFAC SDN, OFAC Consolidated (US Treasury) and UK OFSI sanctions lists. POST {name}. Returns {hit, risk: clear|review|high, matches:[{name, list, program}], lists_screened}. Use for AML/KYC/KYB compliance before onboarding, paying or transacting with a counterparty. Lists refresh every 12h from the official sources.

Search live Amazon products by keyword. Returns normalized ASIN, title, current price/range, currency, rating, review count, bought-past-month signal, canonical product URL, image, rank, Amazon Choice/Best Seller flags and delivery text. Supports location, language and result-depth controls for shopping agents, Amazon product research and ecommerce discovery.

Inspect a public webpage for automation readiness, integration links, forms, robots policy, and security headers.

Fetch the current product price from a public e-commerce product URL and optionally compare it with a target price.

Fetch one public HTTPS URL and return deterministic structured web context: final URL/status, title/description, canonical, headings, readable content, structured-data types, freshness observations, evidence, and methodology.

Kairostamp evidence snapshot: server-side fetch of one public http/https URL, returning a time-stamped receipt with the response status, content type, byte length, SHA-256 of the body, redirect chain, a text preview, and a tamper-evident hash-chain entry.

Run ONE named check against a response you paste

Publisher verification: independent protocol probe and liveness attestation for a submitted URL.

Deterministic URL Context Pack: fetch one public HTTPS page and return bounded title, description, canonical, heading outline, main text, JSON-LD types, freshness observations, and size/token estimates. No LLM. No login. No crawl. Methodolo