Web Scraper

com.agishubv3.0.0更新於 Oct 1, 2026

10 pay-per-call web tools: markdown, CSS scrape, links, crawl, screenshot, PDF, browser.

已驗證Streamable HTTP可網頁執行網頁搜尋與抓取瀏覽器自動化AI 與機器學習

概覽

AI 產生的概覽

遠端 MCP 伺服器，提供網頁擷取、爬取、截圖、PDF 與 AI 工具；MCP 層可免費試用，HTTP 層按次付費。

可以做什麼
將網頁擷取為精簡 Markdown、擷取結構化 JSON、取得連結、對網站建立對應或爬取，並用無頭瀏覽器截圖或產生頁面快照，支援 CSS 選擇器擷取與腳本化瀏覽器步驟。產生 PDF、圖片、語音音訊、QR 碼與短連結，並送出會重試的 webhook。執行 AI 任務：對話、分類、向量化、實體擷取與摘要。換算單位、貨幣與時區，查詢假日、尋找會議時段，並查詢加密貨幣價格。儲存並以語意搜尋持久記憶。
適用情境
適合助理需要讀取或監控公開網頁、把頁面轉成 Markdown 或 PDF、收集網站連結，或在不另外設定 API 金鑰的情況下使用小型 AI 與換算工具時。
執行需求
透過 streamable HTTP 連線的遠端 MCP 端點；未宣告帳號、API 金鑰或環境變數。付費 HTTP 層需要支援 x402 的用戶端，以及持有 Base 上 USDC 的錢包。
安裝前請注意
付費 HTTP 路由每次呼叫都會扣除 Base 上真實且不可逆的 USDC，建議使用小額專用錢包作為支出上限；MCP 層據稱免費。擷取與瀏覽器工具會存取任意公開網址，儲存的記憶與短連結會長期保留。Webhook 會把資料傳送到第三方網址。

安裝

在 SourceWeft 中

  1. 開啟 儀表板中的 Web Scraper，將其新增到工作區。
  2. 為需要使用其工具的對話啟用該服務。

Web executable，透過 Streamable HTTP。 遠端服務在工作區中設定後即可從網頁執行環境執行。

交給代理程式安裝

把這段提示詞貼上到 Claude Code、Codex、Cursor 等代理程式，或 SourceWeft 對話中。代理程式會讀取該伺服器的安裝指南，先向你展示發布者、端點和登入方式，經你同意後再把它新增到你的客戶端。金鑰不會寫進任何命令。

閱讀 https://sourceweft.com/mcp/com.agishub%2Fweb-scraper/install.md，並按其中說明把這個 MCP 伺服器接入我的客戶端。修改任何設定之前，先告訴我它的發布者、端點和登入方式，等我確認。不要把 API 金鑰寫進命令或我會提交的檔案；問我金鑰應從哪裡取得。

檢視代理程式所依據的安裝指南

其他 MCP 客戶端

把它新增到你客戶端的 mcpServers 設定中。

{
  "mcpServers": {
    "web-scraper": {
      "type": "http",
      "url": "https://api.agishub.com/mcp/web"
    }
  }
}

README

AgisHub

Pay-per-call tools for AI agents — no signup, no API keys, just USDC.

▶ Try any tool live in your browser — no install

Live data & utilities over x402 micropayments on Base. Your agent pays a few tenths of a cent per call from its own wallet; no accounts, no monthly plans.

[x402] [Base] [MCP] [License]

Website · Live API · OpenAPI

Why AgisHub

Traditional APIs make an agent stop and ask a human: sign up, create an account, copy an API key, add a credit card, read the docs. AgisHub tools are callable the moment an agent has a funded wallet — the server answers HTTP 402, the agent signs a USDC payment on Base, and the data flows back. Under a second, no human in the loop.

Raw API + key + planSingle‑vendor MCPAgisHub
Sign‑up / API keyrequiredrequirednone
Billingmonthly / creditssubscriptionper call (USDC)
Agent can pay itself✗✗✓ (x402)
Free tier for discoveryvariesvaries✓ (free MCP)
Multiple tools, one door✗one vendor✓

Free MCP tier to discover and prototype; pay‑per‑call x402 HTTP tier for production, priced per request in USDC.

Two doors, one platform

  • MCP (free) — remote Model Context Protocol server. Point Claude / Cursor / any MCP client at the URL and the tools appear.
  • HTTP x402 (paid) — POST /v1/<tool>; unpaid requests get a 402 challenge, the client signs a USDC payment (EIP‑3009 on Base) and retries. USDC contract 0x8335…2913.
MCP (free):   https://api.agishub.com/mcpHTTP (paid):  POST https://api.agishub.com/v1/<tool>Discovery:    https://api.agishub.com/openapi.json

Tools

34 tools across 7 categories. Prices are per call in USDC on Base, from $0.01. Every tool is free to try over MCP; the HTTP endpoint is the one that charges.

🤖 AI & Inference

Language models, embeddings and classification — no external API key.

ToolWhat it doesEndpointCost
ai.chatAsk a general-purpose LLM a question or give it an instruction, with an optional system prompt. No external API key required./v1/chat$0.02
ai.classifyClassify a text into exactly one of the candidate labels you provide (e.g. sentiment, topic, intent)./v1/classify$0.02
ai.embedTurn text into a numeric embedding vector for semantic search, RAG and similarity. Multilingual./v1/embed$0.01
ai.extractExtract named entities from text — people, organizations, locations, dates and miscellaneous — returned as structured JSON./v1/extract$0.02
ai.summarizeSummarize a block of text into a short abstract, with an optional target length. No external API key required./v1/summarize$0.02

🕸️ Search & Web

Read, scrape, crawl and map any public page.

ToolWhat it doesEndpointCost
web.browserDrive a headless browser: open a URL and run an ordered list of steps — click, type, press keys, wait, extract text and screenshot. For flows the plain scraper can't reach (logins, forms, multi-step pages)./v1/browser-automate$0.10
web.crawlCrawl multiple pages of a site, respecting link depth and domain limits. Returns async job_id; results include markdown or HTML per page. Use it to fetch and process many pages of content at once./v1/crawl$0.10
web.extractFetch any public web page and return its main content as clean, token-efficient Markdown (title, description, headings, links, lists). Set render:true to execute JavaScript first for single-page apps or JS-heavy pages that would otherwise come back empty. Built for RAG and for agents that need to read the contents of a URL./v1/web-scraper$0.03
web.extract_structuredAI-powered structured extraction: give a URL plus a natural-language prompt and/or a JSON Schema, and get back clean structured JSON (e.g. product name, price, rating). Renders the page in a headless browser first, so it works on SPAs./v1/extract-structured$0.05
web.linksReturn every hyperlink on a JavaScript-rendered page as a list of absolute URLs, with options to keep only visible links or only same-site links. Backed by a headless browser. Use it to map a site or seed a crawler./v1/links$0.03
web.mapDiscover all URLs reachable from a domain within a link depth limit. Returns a flat list of absolute URLs, respects robots.txt crawl delays. Use it to map a site's structure before crawling./v1/crawl-map$0.05
web.scrapeExtract specific elements from a JavaScript-rendered page by CSS selector. Give a list of selectors (e.g. 'h1', '.price', 'a.product') and get back the text and attributes of every match. Backed by a headless browser, so it works on SPAs and JS-heavy pages./v1/scrape$0.03
web.screenshotCapture a PNG screenshot of any public URL — full page or just the viewport, at a chosen size — returned base64-encoded. Backed by a headless browser./v1/screenshot$0.03
web.snapshotCapture several representations of a page in one call — rendered HTML plus a PNG screenshot by default, and optionally Markdown and the accessibility tree. Backed by a headless browser. Saves round-trips when an agent needs both the content and a visual of a page./v1/snapshot$0.05

📊 Data & Analytics

Time, timezones, units and currency.

ToolWhat it doesEndpointCost
data.convert_unitsConvert a value between units of the same category: length, mass, volume, speed, area, digital storage, time, and temperature (Celsius/Fahrenheit/Kelvin)./v1/units-convert$0.01
data.currency_convertConvert an amount between currencies using live daily exchange rates (ISO 4217 codes, e.g. USD, EUR, GBP, JPY). Returns the converted amount and the rate used./v1/currency-convert$0.01
time.calculateTimezone-aware date arithmetic: add or subtract a duration to a datetime (days are calendar-based and DST-safe; hours and minutes are absolute), or compute the difference between two datetimes that may be in different zones./v1/date-calculate$0.01
time.convertConvert a specific date/time from one IANA timezone to another (single or batch). Accepts ISO 8601 or natural language ('next Tuesday 3pm'). Returns the converted datetime with UTC offset, zone abbreviation and DST flag./v1/time-convert$0.01
time.holidayCheck whether a given date is a public holiday in a country (identified by its ISO 3166-1 alpha-2 code), and return the holiday name if so. Backed by an authoritative public-holiday dataset./v1/time-holiday$0.01
time.meeting_slotsFind working-hour time slots that overlap across participants in different timezones for a meeting of a given duration, excluding weekends and (when a country is given per participant) that person's public holidays. Free tier returns at most 1 slot; the paid endpoint returns every matching slot (see 'upgrade' in the response)./v1/meeting-slots$0.10
time.nowGet the current local time in an IANA timezone, including the UTC offset, zone abbreviation and whether DST is in effect. Use whenever you need to know what time it is 'now' somewhere./v1/time-now$0.01
time.offsetGet the exact UTC offset of an IANA timezone at a given instant, DST-aware. Correctly handles fractional offsets such as India +05:30 and Nepal +05:45./v1/time-offset$0.01
time.timezonesList or search valid IANA timezone identifiers by city/region/country, and resolve a city/country name to its timezone(s). Use it to discover the exact identifier to pass to the other tools./v1/timezones$0.01

💱 Market Data

Live crypto prices.

ToolWhat it doesEndpointCost
market.crypto_priceGet live USD spot prices for one or more cryptocurrencies by ticker symbol (e.g. BTC, ETH, SOL)./v1/crypto-price$0.01

🖼️ Media & Generation

Audio, images and documents.

ToolWhat it doesEndpointCost
audio.speakConvert text into spoken audio (returned base64-encoded MP3), in several languages./v1/audio-speak$0.05
audio.transcribeTranscribe an audio file (given by public URL) to text. Handles mp3, wav, m4a, ogg and more./v1/audio-transcribe$0.05
document.pdfRender a public URL or a raw HTML string into a PDF document, returned base64-encoded. Backed by a headless browser. Use for invoices, reports, receipts and any HTML-to-PDF need./v1/pdf$0.05
image.generateGenerate an image from a text prompt (returned base64-encoded PNG)./v1/generate-image$0.10

🔧 Developer Tools

QR codes, short links and guaranteed webhook delivery.

ToolWhat it doesEndpointCost
qr.generateGenerate a QR code for any text or URL. Returns an inline SVG plus a data URI, with selectable size, quiet-zone margin and error-correction level./v1/qr-generate$0.01
url.shortenShorten a long URL into a compact api.agishub.com/s/ link that redirects to the original. Codes are stored for a year./v1/shorten$0.01
webhook.sendDeliver a webhook (POST/PUT/PATCH a JSON payload to a URL) with guaranteed, retried delivery. Returns immediately with a job_id; AgisHub keeps retrying in the background until it succeeds./v1/webhook-send$0.02
webhook.statusCheck the delivery status of a webhook job (queued / retrying / delivered / failed) by its job_id./v1/webhook-status$0.01

🧠 Knowledge & Memory

Persistent, semantically searchable memory.

ToolWhat it doesEndpointCost
memory.searchSemantically search a memory collection (namespace) and return the most relevant stored entries. The retrieval half of RAG./v1/memory-search$0.01
memory.storeStore a piece of text in a persistent, searchable memory collection (namespace). Embedded and indexed for later semantic recall./v1/memory-store$0.01

Quick start

Use the free MCP tools (Claude Code example):

bash
claude mcp add --transport http timezone https://api.agishub.com/mcp

Pay per call over HTTP x402 (any x402 client / wallet):

bash
# Unpaid → 402 challenge; an x402-aware client signs & retries automatically.curl -X POST https://api.agishub.com/v1/web-scraper \  -H 'content-type: application/json' \  -d '{"url":"https://example.com","render":true}'

Your agent's wallet only needs USDC on Base — the facilitator covers gas.

For agents & LLMs

  • Payments are real USDC on Base mainnet and irreversible. Fund a dedicated agent wallet with a small balance; that balance is your spending cap.
  • The free MCP tier never charges. Only the /v1/* (x402 HTTP) routes take payment.
  • extract fetches arbitrary public URLs; it will not reach private/internal hosts.

Pricing

Per‑call, in USDC on Base — no subscription, no signup, no API key. Five tiers, by what a call actually costs to run:

TierWhatExamples
$0.01Lookups, conversions, memorytime.now, data.currency_convert, memory.search
$0.02LLM inference, guaranteed deliveryai.chat, ai.classify, webhook.send
$0.03Browser, one pageweb.scrape, web.extract, web.screenshot
$0.05Heavy rendering, mediaweb.snapshot, document.pdf, audio.transcribe
$0.10Orchestration, generationweb.crawl, web.browser, image.generate

Roughly: $1 ≈ 100 timezone calls, ≈ 33 web extractions, or 10 crawls of up to 100 pages each. Settlement fees on Base are a fraction of a cent and paid by the facilitator.

web.crawl and web.map are capped per call (100 pages, 200 URLs) and priced flat, so a larger site is several calls at the same rate rather than one unbounded charge.

License

MIT © AgisHub

來源：README.md，提交 0fbbc2d

工具

0
SourceWeft 尚未讀取此伺服器的工具。

版本歷史

1
  1. v3.0.0最新Sep 16, 2026

評分

已滿足適用檢查項的 50%

該評分衡量此收錄條目的完整程度，依據註冊中心條目、端點與公開來源計算得出。它不是對伺服器本身的評價，也不表示伺服器安全或目前可用：可連通性是帶日期的檢查結果，評分也不包含安裝量或使用者評價。

  • 近 30 天內可連通必需25 分

    請保持端點能夠回應 MCP 握手；SourceWeft 每 30 天探測一次。

  • 已宣告認證方式必需10 分

    請在註冊中心條目中宣告憑證請求標頭，或保持端點能夠回應探測，以便判定其認證方式。

  • 已列出工具15 分

    請讓伺服器在無憑證時回應 tools/list，或發布可在沙箱中無憑證啟動的軟體包。

  • 描述必需15 分
  • 圖示10 分
  • 文件15 分
  • 授權條款10 分