OCR MCP Server
OCR MCP Server
基于 Claude Vision 能力的文档文字识别 MCP 工具,无需 Tesseract/PaddleOCR 等第三方 OCR 引擎,直接利用大模型多模态视觉能力提取文字。
✨ 特性
零 OCR 引擎依赖 — 不安装 Tesseract、PaddleOCR,全部识别能力来自 Claude Vision API
三种工具 — 图片识别、智能 PDF 识别、强制 OCR 扫描件
智能降级 — 文字版 PDF 直接提取内嵌文字,零 API 成本;扫描件自动转图片走 Vision
结构化输出 — 支持 Markdown 表格、JSON、纯文本,适配表格/表单/手写体等场景
即插即用 — 配置 Claude Desktop 后,自然语言直接调用
Related MCP server: mcp-upstage
🛠 工具
工具 | 用途 |
| 识别图片文字,支持印刷体、手写体、表格,输出 Markdown/JSON |
| 智能处理 PDF:文字版直接提取内嵌文字(省 token),扫描件自动走 Vision OCR |
| 强制全页 Vision OCR,适配纯扫描件/手写 PDF |
📦 安装
# 克隆仓库
git clone https://github.com/你的用户名/ocr-mcp-server.git
cd ocr-mcp-server
# 安装依赖(仅 mcp / anthropic / PyMuPDF)
pip install -r requirements.txt
# 设置 Anthropic API Key
export ANTHROPIC_API_KEY="sk-ant-xxxxx"⚙️ 配置 Claude Desktop
编辑 claude_desktop_config.json:
macOS:
~/Library/Application Support/Claude/Windows:
%APPDATA%\Claude\
{
"mcpServers": {
"ocr": {
"command": "python",
"args": ["/你的绝对路径/ocr-mcp-server/server.py"],
"env": {
"ANTHROPIC_API_KEY": "sk-ant-xxxxx"
}
}
}
}重启 Claude Desktop 即可使用。
🚀 使用
在 Claude Desktop 中直接用自然语言触发:
"帮我识别这张图片里的文字" → 自动调用
recognize_image"提取这个 PDF 里的表格" → 自动调用
recognize_pdf"这个扫描件里写了什么" → 自动调用
recognize_pdf_ocr
🧠 工作原理
图片 ──→ base64 编码 ──→ Claude Vision API ──→ 结构化文本
PDF ──→ PyMuPDF 检测文字层
├─ 有文字层 → 直接提取(零成本)
└─ 扫描件 ──→ 转图片 ──→ base64 ──→ Vision API核心思路:用 LLM 的视觉理解能力替代传统 OCR 管线,对复杂排版、表格、手写体、印章等场景的理解更强,且能保持语义结构。
🔧 技术栈
📄 License
MIT
This server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Related MCP Servers
- FlicenseNot gradedqualityDmaintenanceEnables Claude to read and analyze PDF documents with automatic OCR processing for scanned files. Features intelligent text extraction, caching for performance, and secure file access with search capabilities.1

mcp-upstageofficial
AlicenseBqualityCmaintenanceEnables AI assistants to extract and structure content from documents (PDFs, images, Office files) via Upstage AI's APIs, with seamless Claude Desktop integration.213MIT
@parserelay/mcpofficial
AlicenseAqualityBmaintenanceEnables document parsing into structured, confidence-scored fields via the scan tool, working with any MCP host like Claude Desktop or Cursor.188MIT- FlicenseNot gradedqualityDmaintenanceEnables OCR on images and PDFs, including full-page OCR, region OCR by description or bounding box, and caching with summary capabilities.
Related MCP Connectors
Turn any PDF into structured JSON via AI + OCR: invoices, bank statements, contracts.
Read PDFs and images as markdown or text, with exact costs and hard spend caps. $0.75/1k pages.
OCR, transcription, file extraction, and image generation for AI agents via MCP.
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/zhang2gongzi/ocr-mcp-server'
If you have feedback or need assistance with the MCP directory API, please join our Discord server