pdf-agent-mcp
pdf-agent-mcp
中文
pdf-agent-mcp 是一个本地 MCP 服务,给 AI agent 提供 PDF 文本层读取能力。
工具列表
inspect_pdf:检查 PDF 基本信息、页数、是否可能有文本层extract_pdf_text:按raw/lines/blocks抽取文本extract_pdf_outline:提取 PDF 目录(书签)extract_pdf_page:提取单页文本项和坐标
使用说明
环境要求:Node.js 22+
npm install
npm run devnpm run lint
npm test
npm run build推荐直接用 npx 启动:
npx -y github:sanhua1/pdf-agent-mcpAgent 自然语言交互示例
在 Claude/Codex 里可直接说:
先帮我 inspect 这个 PDF:/path/to/doc.pdf把 1-5 页按 lines 模式提取出来第 10 页排版乱,改用 blocks 模式再提取一次先读取目录,再按章节整理成 Markdown 摘要
Claude Code 配置方法
{
"mcpServers": {
"pdf-agent-mcp": {
"command": "npx",
"args": ["-y", "github:sanhua1/pdf-agent-mcp"]
}
}
}Codex 配置方法
[mcp_servers.pdf-agent-mcp]
command = "npx"
args = ["-y", "github:sanhua1/pdf-agent-mcp"]English
pdf-agent-mcp is a local MCP server for extracting text-layer content from PDF files.
Tools
inspect_pdf: inspect metadata, page count, and text-layer hintsextract_pdf_text: extract text inraw/lines/blocksmodesextract_pdf_outline: read PDF bookmarks/outlinesextract_pdf_page: extract text items with coordinates from a single page
Quick Start
Requirement: Node.js 22+
npm install
npm run devRun with npx:
npx -y github:sanhua1/pdf-agent-mcpLatest 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/sanhua1/pdf-agent-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server