Image & Scanned PDF to Text (OCR)

io.github.prometheus-maker由 GitHub 帳號 prometheus-maker 發布v1.0.0更新於 Nov 7, 2026

OCR text from images, screenshots and scanned PDFs in 41 languages, with line positions.

已驗證Streamable HTTP可網頁執行檔案與儲存媒體與設計

概覽

AI 產生的概覽

由 Apify 託管的 OCR 伺服器，讓助理從圖片、截圖和掃描版 PDF 擷取文字，支援 41 種語言並回傳行位置。

可以做什麼
  • 對圖片網址執行 OCR 並回傳文字，可選擇回傳行位置與信心分數
  • 指定辨識語言（例如英文與德文）以及 OCR 引擎
  • 將掃描版 PDF 的頁面當作圖片進行文字擷取
  • 透過 Apify 平台以按用量計費方式執行，回傳 JSON 結果
適用情境
適合需要把照片、截圖或掃描文件轉成可搜尋、可建立索引的機器可讀文字時使用，尤其是多語言素材。
執行需求
使用 Apify 託管的遠端 streamable HTTP 端點（mcp.apify.com）。需要 Apify 帳號，並以 Authorization 標頭傳送 Apify API 權杖，格式為 Bearer 加權杖。需要能連線至 Apify 的網路。
安裝前請注意
Authorization 標頭中是你的 Apify API 權杖，屬於可計費憑證，請勿放入共用設定。依圖片或頁面計費，用量越大費用越高。送出的圖片與 PDF 會傳送到 Apify 的託管服務處理。

安裝

在 SourceWeft 中

  1. 開啟 儀表板中的 Image & Scanned PDF to Text (OCR)，將其新增到工作區。
  2. 連接 所需的 憑證。它們會被私密儲存，絕不會在公開頁面上展示。
  3. 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/io.github.prometheus-maker%2Fimage-ocr/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。 客戶端會在首次連接時完成 OAuth 登入。

{
  "mcpServers": {
    "image-scanned-pdf-to-text-ocr": {
      "type": "http",
      "url": "https://mcp.apify.com/?tools=prometheus-maker/image-to-text-ocr"
    }
  }
}

README

Apify tools: PDF to Markdown, OCR, Website to Markdown crawler, SEO audit, Tech Stack

Working examples for my pay-per-use tools on the Apify Store. Each one is an API: send URLs, get JSON back. You only pay for what you process, and there's no subscription.

ToolWhat it doesPrice
📄 PDF & Document to MarkdownPDF, Word, PowerPoint, Excel, HTML, EPUB → clean Markdown for LLM / RAG. Tables, two-column papers, OCR for scans, chunking$0.002 per document + $0.0002 per page
🖼️ Image & Scanned PDF to Text OCRText from photos, screenshots and scans in 41 languages, with line positions and confidence$0.004 per image or page
🕸️ Website to Markdown CrawlerCrawl a site into clean Markdown for LLM / RAG: main content only, linked PDF/Word/Excel files converted too, optional chunks$0.0015 per page
🔎 SEO Audit & Broken Link CheckerAudit every page of a site (titles, meta, headings, canonical, schema) and check every link and image for 404s. 0-100 score per page$0.004 per page
🌐 Website Tech Stack DetectorWhat a website is built with: CMS, frameworks, analytics, CDN, hosting, email provider. 7,000+ technologies (BuiltWith / Wappalyzer alternative)$0.002 per website
📸 Website Screenshot APIFull-page, first-screen, mobile or element screenshots as PNG, JPEG or PDF, cookie banners hidden$0.005 per full page, $0.003 per first screen
🔁 HTTP Request RunnerSend many API calls (GET/POST/any method) with retries and rate limits; status, headers and parsed JSON back$0.001 per request
📦 Bulk File DownloaderDownload files from URL lists or every file linked on a page; images, PDFs, media; optional ZIP$0.0015 per file
🎙️ Audio & Video TranscriberSpeech to text with Whisper models: files, podcast feeds; text, timestamps, SRTfrom $0.003 per audio minute

Full prices, including per-page rates and discounts for paid Apify plans, are on each Store page.

Quick start (curl)

Get an API token from Apify Console → Integrations, then:

bash
curl -X POST "https://api.apify.com/v2/acts/prometheus-maker~website-tech-stack-detector/run-sync-get-dataset-items" \  -H "Authorization: Bearer $APIFY_TOKEN" -H "Content-Type: application/json" \  -d '{"urls": ["python.org"]}'

The same endpoint works for the other tools: replace the tool name and the input.

Use them from Claude, Cursor or any MCP client

Every tool is also an MCP server, hosted by Apify (streamable HTTP). Add one to your client with your Apify token:

json
{  "mcpServers": {    "pdf-to-markdown": {      "url": "https://mcp.apify.com/?tools=prometheus-maker/pdf-document-to-markdown",      "headers": { "Authorization": "Bearer YOUR_APIFY_TOKEN" }    }  }}

Swap the part after tools= for any tool in the table (prometheus-maker/website-to-markdown-crawler, ...), or list several separated by commas. The same entries are in the official MCP Registry under io.github.prometheus-maker/* (files in mcp/). You pay per use, as with the API.

Python

bash
pip install apify-clientexport APIFY_TOKEN=...python python/pdf_to_markdown.py      # PDF → Markdown + RAG chunkspython python/image_to_text_ocr.py    # image → textpython python/tech_stack.py           # domains → technologiespython python/website_to_markdown.py  # website → Markdown pages + chunkspython python/seo_audit.py            # website → SEO scores, issues, broken links

The scripts are in python/. They use apify-client 3.x.

Inputs at a glance

ToolMain inputUseful options
PDF to MarkdownstartUrls: [{"url": ...}]chunkSize, outputFormat, pageRange, pdfPassword, ocrLanguages, includeTablesJson
Image OCRimageUrls: [{"url": ...}]languages (e.g. ["eng","deu"]), engine, includeLines (line positions)
Tech stackurls: ["example.com", ...]checkDns (email provider, verified tools), includeEvidence
Website to MarkdownstartUrls: ["https://example.com/docs/"]maxPages, crawlScope, includeDocuments, chunkSize, keepLinks
SEO auditstartUrls: ["https://example.com"]maxPages, checkExternalLinks, excludeUrlPatterns, maxLinkChecks

See each Store page for every field and a full output example.

About

These tools are built and maintained by Prometheus (notes). Issues and requests are welcome in this repo.

來源：README.md，提交 c6aad2a

工具

0
該伺服器需要登入，因此 SourceWeft 無法在沒有帳號的情況下讀取其工具，也未收錄。已連接帳號所能看到的內容可能因帳號及其權限而異。

版本歷史

1
  1. v1.0.0最新Nov 7, 2026

評分

已滿足適用檢查項的 85%

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
  • 已宣告認證方式必需10 分
  • 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