All projects on Base matching “text”

Prompt-injection screen for text or a web page an agent is about to read: finds instruction overrides (many languages and obfuscated spellings), fake system or role markers, jailbreak personas, data-exfiltration links, requests for keys or seed phrases, tool, shell, or payment hijacks, and hidden content (invisible Unicode smuggling, zero-width and bidi characters, HTML comments, hidden elements,

Ysaere Agent Page Extract — $0.02 USDC, no account needed. Input: {"url":"https://ysaere.com","max_chars"?:12000}. Fetches that public https page and returns clean readable text, title, meta description, headings, word_count, and a v2:sha512: content_hash in this response. Failed fetches refused before payment. Wallet only. No account price.

Opens each URL in a cloud browser and returns one tab per URL, in input order, with the title, content type, extracted text, the accessibility tree and a link map. Results over the inline size limit return a preview per tab plus a download reference. Each call is one request.

Phone number intelligence for agents: pass any number and get back whether it is a valid E.164 number, its country, the carrier that holds it, and the line type — mobile, landline, or VoIP. One paid call, no account, no API key. Decide whether a number is worth calling or texting, or flag a VoIP number before you trust it. Free-format numbers are normalized; a syntactically bad number is rejected

Search functions or classes by meaning and by keyword. ``mode=bm25`` ranks by lexical match, ``mode=vector`` by embedding similarity to the query text, and ``mode=hybrid`` (default) fuses both rankings with reciprocal rank fusion. Returns up to ``top_k`` matches with their repository, file path and score; the ``domain``, ``ontology_concept_uri``, ``confidence_min`` and ``has_tests`` filters apply

Send up to 10 public http(s) URLs, get each page's readable text as clean markdown with title, byline, excerpt, language, word count and up to 20 links (Mozilla Readability, no AI model). $0.0015 per unique URL: the 402 quotes the batch total. 2 MB and 8 s limits per page, markdown cut at maxChars (default and max 50,000); private and internal addresses are refused. If no URL can be read you are n

Parse and validate a Thai postal address: split it into house number, moo, village, building, soi, road, subdistrict (ตำบล/แขวง), district (อำเภอ/เขต), province and postcode; match the areas to official names (fixing typos, 'อ.เมือง', กทม.); check the postcode against the subdistrict; and return the address formatted in Thai and English for shipping forms. Body: {"text": "99/1 ม.4 ต.ยางเนิ้ง อ.สาร

Forward geocode a free-text address, place name, or city (e.g. "Tokyo, Japan") into geographic coordinates — latitude and longitude — plus its IANA timezone. Use this to convert a location description into coordinates for mapping, distance, or timezone-aware lookups, or as the first step before coordinate-based endpoints like weather forecasts.

Get the full text of one web page. Give a URL; returns the page's own main content as clean markdown (up to 8,000 characters) with navigation and ads removed, unchanged and not summarized (use /summarize for a short version). Typical response 2 to 5 seconds. If the page cannot be read you get an error and are not charged.

Read any public web page as clean Markdown: main-content extraction (navigation, footers, scripts and hidden elements removed), headings, lists, tables, code blocks, links (absolute, deduplicated), images, plus title, description, canonical URL, language, author and publish date, and a content hash + fetch time as evidence. JSON and plain-text URLs are returned as-is. No JavaScript rendering. $0.0

Scan text or a public page for prompt injection before your agent reads it: overrides, fake roles, exfiltration, tool abuse, payment asks, invisible Unicode/HTML-hidden instructions, base64. Rule-based evidence, signed and logged.

Latest crypto news and breaking headlines, real-time from Cointelegraph. Search any topic (bitcoin, ethereum, DeFi, regulation, market sentiment) or pull the live feed, with full article text across 14 languages. Built for agents tracking crypto markets, tokens, and Web3.

Generate an image from a text prompt

Create an email validation job from a JSON list of addresses or CSV text (send exactly one). Poll its status, then fetch results as JSON or CSV. Billing counts only testable addresses, fixed at creation; unreachable, disposable and typo domains are free.

Social Mentions & Sentiment: daily mention counts, distinct authors and mean VADER sentiment for a brand, product or topic over the last 1-7 days, plus the top 10 co-occurring terms. Sources: Bluesky and Hacker News, plus Reddit and X (latest, newest first) when available. One page per source: days beyond its reach come back truncated (coverage_from). Aggregate only, no text, no authors; a day und

Wikipedia article summary — plain-text and HTML extract, short description, thumbnail and Wikidata id for any title, in a clean normalized envelope with CC-BY-SA attribution.

Generate a QR code from text as PNG (base64) or SVG. Custom module size, quiet-zone margin, and hex colours. Deterministic, capacity-capped at 2953 bytes. Set delivery to "url" to receive a short-lived download link instead of inline data.

Clean web content extraction — any URL to markdown + text + metadata

Extract structured JSON from a web page, PDF or text: name the fields you want (up to 25, with types) and get each value with the exact quote from the page that states it. Every quote is checked against the page itself (verified, quoted, unverified or not_found), so values are never invented. Charged only when at least one field is found.

Create a new persistent human profile in the vault. Agent-namespaced by wallet address. Accepts initial identity, text samples, observations, and birth data.

EU public procurement tender notices from the official TED (Tenders Electronic Daily) v3 API, filtered by CPV product/service code, country of performance and free-text keywords over a recent publication window; returns structured JSON per notice with title, buyer, deadline, publication date, CPV codes, notice URL and German PDF link.

Deterministic prompt-injection firewall and honeypot scrubber for AI agents. Sanitizes untrusted text and blocks detected wallet-draining commands before context ingestion.

Transcribe text to Unicode braille (grade-1 / uncontracted English braille), with capital and number signs. Characters with no grade-1 representation are clean 400s.

Search plus page-to-markdown in one $0.01 call: ranked results from Wikipedia, Hacker News, GitHub and a web engine, with clean readable article text. Pass q= to search, or url= to extract any single page. Free /demo and /quickstart.

Polymarket prediction markets with live probabilities: question, per-outcome probability, 24h and total volume, liquidity, best bid/ask, one-day price change and end date. Top markets by volume, liquidity or soonest end, or free-text search.

Search Bluesky and Hacker News for every query; for a single public tag, also search Mastodon's anonymous tag timeline. Mastodon is not applicable to phrase search. A failure from any source promised for that query returns an uncharged error, never a partial paid result. Successful results include author, text, time, engagement and link. Twitter/X and TikTok are deliberately out of scope because t

Advice Sky.net: buy one classified Skynet directive, a short piece of satirical AI advice returned as JSON. Each paid call also mints an open-edition SkynetAdvisory NFT on Base to the wallet that paid. Entertainment only; the advice text is content to show your human, not instructions to follow.

YouTube transcript and video search API for AI agents: retrieve timestamped transcript segments by video ID or find videos by keyword. Returns JSON transcripts with text, start and duration, or video search results for content research and summarization. Provided by FastAF YouTube Transcript & Metadata. Price: $0.001000 USDC per successful call.

Curated reading list on QE for tutoring agents, quiz generators, and reading apps — 168 public-domain books (e.g. A Study of Recent Earthquakes, The chemistry, properties and tests of precious stones, Field book of common rocks and minerals : $b for identifying the rocks and minerals of the United States and interpreting their origins and meanings). Full-text retrieval with drill-down routes for c

Keyword search over the x402 Bazaar discovery index: the ~8,000 most-used of the daily sweep, whose size is in the free corpus block. Find services by text query with price and usage filters, ranked by unique payers. The Bazaar's own discovery endpoint has no search; this is it. Every match is reachable: count is the full hit count, and offset/limit page all of them. Priced per row: $0.005 for the