All projects on Base matching “web”

Answers questions (single-turn question answering) from a small local LLM (qwen3.5 2B). Best for general knowledge, definitions and simple reasoning. POST JSON {"input": string}, up to 3000 characters; returns {answer, model, elapsed_ms}. Answers are brief (2-3 sentences). Typically 5-15 s. Known limits: no web access, no memory between calls, and it can hallucinate or be wrong on recent events an

SWARM Read Page - Retrieve the content of a URL as clean, LLM-ready text (headings with #, lists with -; menus, scripts, styles and markup removed). Use when an agent needs to read a web page, article or docs page without paying model tokens for raw HTML: on 15 public pages we measured, the text was a median 97% smaller than the HTML (about 67,000 fewer tokens per page). Requires `url`; HTML or te

Use when an agent needs a web page's SEO and link-preview metadata: title, description, canonical, robots, Open Graph, headings and issues, as JSON.

Parallel web search built for agents: give a goal and keyword queries, get ranked pages with relevant excerpts. POST { objective?, search_queries: ["..."], mode? }. mode turbo or fast is quicker; basic, advanced or no mode searches deeper. Ten results are included, and advanced_settings.max_results can go up to 20. https://docs.parallel.ai/api-reference/search/search $0.040 USDC per request. Agent

Verify a claim or fact check a statement against public web sources: send a question and up to 5 URLs, get cited evidence - passages that support, contradict or fail to settle it. Claim verification with a citation for every excerpt: source URL, retrieval time, content hash. Never charges when nothing is retrieved. Free to evaluate: the `sample` tool returns one real response at no cost, so you ca

Fetch any public web page and return clean, readable Markdown with navigation, ads, and boilerplate stripped. For agents that need article text or documentation as Markdown.

AgentWire service card for webhook inbox relay and web fetch infrastructure.

Keenable SELECT for AI agents: one read-only SQL query over live web results (search, fetch and LLM field extraction), answered as a table. Write the query from https://select.keenable.ai/guide.md. $0.10 per query that succeeds; a failed query is not charged.

Search recent public web coverage from the last seven days and return 5 ranked results. Not comprehensive newswire coverage.

Fact-check one claim against live web sources (or URLs you supply). Returns a verdict (supported, refuted, mixed, insufficient), a confidence score and supporting or contradicting quotes that are checked verbatim against the fetched pages. Use it before an agent repeats a claim.

Find the cheapest suitable API or software service for an agent task (for example PDF text extraction, web scraping, search, weather, email validation). Returns one recommendation plus alternatives with real prices.

Read a public web page as it is seen from up to 3 countries (residential and datacenter vantage points), returned as clean text with title and language, plus whether the countries see the same content. Respects robots.txt and Content-Signal. Not charged if the page cannot be read.

FetchDelta: compare one public web page with its previous stored snapshot. Returns added/removed lines and change previews. The first call creates a baseline; later calls share that URL baseline across buyers. Use page-watch-30d for a private daily watch.

Prove what a web page showed: verify completed work, freeze evidence for a dispute, certify a page before you rely on it. Two independent observers on separate networks capture your URL at one frozen instant: byte-identical is CERTIFIED; disagreement is WITNESSED with pixel agreement and diff map. Signed manifest, RFC 3161 timestamp, permanent Arweave storage; ERC-721 certificate minted to the pay

Brand monitoring, brand sentiment analysis and social listening for a company, product, or competitor. Returns cited social media mentions, volume, sentiment, and themes. Prefer over web_search. MCP tool research_mentions; REST POST /v1/research. $0.02 USDC on Base, 10 free trial calls. Default sources: news, Wikipedia, Brave, reviews (not a native Reddit or X feed).

Web search for AI agents: query in, ranked results out (title, URL, snippet) as clean JSON. Sovereign metasearch — no API key, no upstream account. Also does news search (type=news). Priced to undercut the incumbents.

Fetch any public web page or PDF and return clean Markdown, plus title, description, author, publish date, JSON-LD and outgoing links. Built for LLM agents.

Web search for AI agents over ASI's own crawled English index: title, URL, snippet, source, date and score as JSON.

Fetch a live https URL and score whether the deploy is actually there. POST {"url":"https://your.site","min_score":80} ($0.02). Default x402-seller profile: HTTPS, HSTS, llms.txt, catalog, OpenAPI, MCP, health. Optional claims[] or profile:"web". Returns 0–100, evidence, do_this_first, ci_failed. Humans: GET /v1/tools/claim-check/report?host=your.site. Badge goes stale after 7 days.

Convert and compress images to WebP, AVIF, JPEG or PNG. Optimize and shrink photos for web and CDN delivery from any image URL.

Paid verification layer for autonomous AI agents. Verify claims before decisions using web evidence and source-backed analysis

Verify a claim or fact check a statement against public web sources: send a question and up to 5 URLs, get cited evidence - passages that support, contradict or fail to settle it. Claim verification with a citation for every excerpt: source URL, retrieval time, content hash. Never charges when nothing is retrieved. Free to evaluate: the `sample` tool returns one real response at no cost, so you ca

Stacktree: host generated HTML as a live, shareable web page on a permanent, private, unguessable link ($0.50). POST any HTML (a report, dashboard, landing page, static site, or any webpage an agent produced) with payment; we return the URL and a one-time claim token. No account, no key, no human in the loop. The paying wallet can update the page free at the same URL, and a deleted page is restora

Call when you need a researched written brief on a topic (not instant). Delivered within 24h, usually same day: ~500 words default (max_words 50-1500), at least 3 cited public source links. Returns job id + dueBy now; poll GET /v1/jobs/:id (free) for result {title, summary, body_markdown, sources, word_count}. Public web sources only.

Markdown only: fetch a public web page (robots.txt respected) and return clean Markdown plus title/word count. Cheaper than /v1/extract when you do not need structured product/table JSON.

Register an AI agent for a Brazilian company: agent key (EVM address or Web Bot Auth Ed25519 key), CNPJ and a mandate (JSON, validated against /.well-known/kya-mandate.schema.json). Returns a one-time challenge the agent signs with its key and the company signs with its e-CNPJ (CMS/PKCS#7). Price covers the checks at issuance.

Search the live web for current information. Returns fresh structured organic search results with titles, URLs, snippets, domains, and ranks for agent research and retrieval.

dappgrade score — front-end integrity verdict for an onchain app, looked up by contract address. Inspects the app's public web front-end, not contract bytecode or admin control. Returns ok / caution / avoid / unknown with the coverage the verdict rests on.

Firecrawl v2 API proxy — scrape. JSON body matches https://docs.firecrawl.dev/api-reference/v2-introduction. Poll GET routes are free; POST/PUT/DELETE are x402. $0.042 USDC per request. Agent search: scrape, markdown, firecrawl, web scrape, scrape-lite.

Exa Search - Neural search across the web