#!/usr/bin/env node
import { Server } from "@modelcontextprotocol/sdk/server/index.js";
import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js";
import {
CallToolRequestSchema,
ListToolsRequestSchema,
} from "@modelcontextprotocol/sdk/types.js";
import { ArXivClient } from '@agentic/arxiv';
import axios from "axios";
import * as fs from "fs";
import * as path from "path";
import { PdfReader } from "pdfreader";
import { tmpdir } from "os";
import { JSDOM } from "jsdom";
// 初始化 ArXiv 客户端
const arxivClient = new ArXivClient({});
// 创建 MCP 服务器
const server = new Server(
{
name: "arxiv-paper-mcp",
version: "1.1.0",
},
{
capabilities: {
tools: {},
},
}
);
// 工具函数：搜索 arXiv 论文
async function searchArxivPapers(query: string, maxResults: number = 5): Promise<{totalResults: number, papers: any[]}> {
try {
const results = await arxivClient.search({
start: 0,
searchQuery: {
include: [
{ field: "all", value: query }
]
},
maxResults: maxResults
});
const papers = results.entries.map(entry => {
const urlParts = entry.url.split('/');
const arxivId = urlParts[urlParts.length - 1];
return {
id: arxivId,
url: entry.url,
title: entry.title.replace(/\s+/g, ' ').trim(),
summary: entry.summary.replace(/\s+/g, ' ').trim(),
published: entry.published,
authors: entry.authors || []
};
});
return {
totalResults: results.totalResults,
papers: papers
};
} catch (error) {
console.error("搜索 arXiv 论文时出错:", error);
throw new Error(`搜索失败: ${error instanceof Error ? error.message : String(error)}`);
}
}
// 工具函数：检查是否有 HTML 版本并获取内容
async function getArxivHtmlContent(arxivId: string): Promise<string | null> {
try {
const cleanArxivId = arxivId.replace(/v\d+$/, '');
const htmlUrl = `https://arxiv.org/html/${cleanArxivId}`;
console.log(`尝试获取 HTML 版本: ${htmlUrl}`);
const response = await axios({
method: 'GET',
url: htmlUrl,
timeout: 20000,
headers: {
'User-Agent': 'Mozilla/5.0 (compatible; ArXiv-Paper-MCP/1.0)'
}
});
// 检查响应状态和内容类型
if (response.status === 200 && response.headers['content-type']?.includes('text/html')) {
const html = response.data;
// 简单检查是否是有效的论文HTML（而不是错误页面）
if (html.includes('ltx_document') || html.includes('ltx_page_main') || html.includes('ltx_abstract')) {
console.log(`成功获取 HTML 版本: ${htmlUrl}`);
return html;
}
}
console.log(`HTML 版本不可用或无效: ${htmlUrl}`);
return null;
} catch (error) {
console.log(`HTML 版本获取失败，将使用 PDF: ${error instanceof Error ? error.message : String(error)}`);
return null;
}
}
// 工具函数：从 HTML 中提取文本内容
function extractTextFromHtml(html: string): string {
try {
const dom = new JSDOM(html);
const document = dom.window.document;
// 移除脚本和样式标签
const scripts = document.querySelectorAll('script, style');
scripts.forEach(el => el.remove());
// 获取主要内容区域
let mainContent = document.querySelector('.ltx_page_main') ||
document.querySelector('.ltx_document') ||
document.querySelector('body');
if (!mainContent) {
throw new Error('无法找到主要内容区域');
}
// 提取文本内容
let text = mainContent.textContent || '';
// 清理文本：移除多余的空白字符
text = text.replace(/\s+/g, ' ').trim();
if (text.length < 100) {
throw new Error('HTML 文本内容过少');
}
return text;
} catch (error) {
console.error("HTML 文本提取失败:", error);
throw new Error(`HTML 解析失败: ${error instanceof Error ? error.message : String(error)}`);
}
}
// 工具函数：获取 AI 领域最新论文
async function getRecentAIPapers(): Promise<string> {
try {
const url = 'https://arxiv.org/list/cs.AI/recent';
console.log(`正在获取 AI 领域最新论文: ${url}`);
const response = await axios({
method: 'GET',
url: url,
timeout: 30000,
headers: {
'User-Agent': 'Mozilla/5.0 (compatible; ArXiv-Paper-MCP/1.0)'
}
});
return response.data;
} catch (error) {
console.error("获取最新 AI 论文时出错:", error);
throw new Error(`获取最新论文失败: ${error instanceof Error ? error.message : String(error)}`);
}
}
// 工具函数：获取 arXiv PDF 下载链接
function getArxivPdfUrl(input: string): string {
try {
let arxivId: string;
let pdfUrl: string;
if (input.startsWith('http://') || input.startsWith('https://')) {
const urlParts = input.split('/');
arxivId = urlParts[urlParts.length - 1];
pdfUrl = input.replace('/abs/', '/pdf/') + '.pdf';
} else {
arxivId = input;
pdfUrl = `http://arxiv.org/pdf/${arxivId}.pdf`;
}
return pdfUrl;
} catch (error) {
console.error("获取 PDF 链接时出错:", error);
throw new Error(`获取PDF链接失败: ${error instanceof Error ? error.message : String(error)}`);
}
}
// 工具函数：下载临时 PDF 文件
async function downloadTempPdf(pdfUrl: string): Promise<string> {
try {
console.log(`正在下载临时 PDF: ${pdfUrl}`);
const response = await axios({
method: 'GET',
url: pdfUrl,
responseType: 'stream',
timeout: 30000,
headers: {
'User-Agent': 'Mozilla/5.0 (compatible; ArXiv-Paper-MCP/1.0)'
}
});
// 创建临时文件路径
const tempPath = path.join(tmpdir(), `arxiv_temp_${Date.now()}.pdf`);
const writer = fs.createWriteStream(tempPath);
response.data.pipe(writer);
return new Promise<string>((resolve, reject) => {
writer.on('finish', () => {
console.log(`临时 PDF 下载完成: ${tempPath}`);
resolve(tempPath);
});
writer.on('error', (error) => {
console.error(`临时 PDF 下载失败: ${error}`);
if (fs.existsSync(tempPath)) {
fs.unlinkSync(tempPath);
}
reject(error);
});
});
} catch (error) {
console.error("下载临时 PDF 时出错:", error);
throw new Error(`下载失败: ${error instanceof Error ? error.message : String(error)}`);
}
}
// 工具函数：提取 PDF 文本内容
async function extractPdfText(pdfPath: string): Promise<string> {
return new Promise<string>((resolve, reject) => {
const texts: string[] = [];
new PdfReader().parseFileItems(pdfPath, (err, item) => {
if (err) {
console.error("PDF 解析失败:", err);
reject(new Error("PDF 解析失败: " + err));
} else if (!item) {
// 解析结束，拼成一段文本
let text = texts.join(' ').replace(/\s+/g, ' ').trim();
if (text.length < 100) {
reject(new Error("PDF 文本提取失败或内容过少"));
} else {
resolve(text);
}
} else if (item.text) {
texts.push(item.text);
}
});
});
}
// 工具函数：解析论文内容（优先 HTML，回退 PDF）
async function parsePaperContent(input: string, paperInfo?: any): Promise<{content: string, source: 'html' | 'pdf'}> {
let tempPdfPath: string | null = null;
try {
// 获取 arXiv ID
let arxivId: string;
if (input.startsWith('http://') || input.startsWith('https://')) {
const urlParts = input.split('/');
arxivId = urlParts[urlParts.length - 1];
} else {
arxivId = input;
}
// 首先尝试获取 HTML 版本
console.log("尝试获取 HTML 版本...");
const htmlContent = await getArxivHtmlContent(arxivId);
let paperText: string;
let source: 'html' | 'pdf';
if (htmlContent) {
// 使用 HTML 版本
console.log("使用 HTML 版本解析内容");
paperText = extractTextFromHtml(htmlContent);
source = 'html';
} else {
// 回退到 PDF 版本
console.log("HTML 版本不可用，回退到 PDF 版本");
const pdfUrl = getArxivPdfUrl(input);
tempPdfPath = await downloadTempPdf(pdfUrl);
paperText = await extractPdfText(tempPdfPath);
source = 'pdf';
}
// 构建输出内容
let outputContent = '';
if (paperInfo) {
outputContent += `=== 论文信息 ===\n`;
outputContent += `标题: ${paperInfo.title}\n`;
outputContent += `arXiv ID: ${arxivId}\n`;
outputContent += `发布日期: ${paperInfo.published}\n`;
outputContent += `内容来源: ${source.toUpperCase()}\n`;
if (paperInfo.authors && paperInfo.authors.length > 0) {
outputContent += `作者: ${paperInfo.authors.map((author: any) => author.name || author).join(', ')}\n`;
}
outputContent += `摘要: ${paperInfo.summary}\n`;
outputContent += `\n=== 论文内容 ===\n\n`;
} else {
outputContent += `=== 论文内容 (来源: ${source.toUpperCase()}) ===\n\n`;
}
outputContent += paperText;
return { content: outputContent, source };
} catch (error) {
console.error("解析论文内容时出错:", error);
throw new Error(`论文内容解析失败: ${error instanceof Error ? error.message : String(error)}`);
} finally {
// 清理临时 PDF 文件
if (tempPdfPath && fs.existsSync(tempPdfPath)) {
try {
fs.unlinkSync(tempPdfPath);
console.log(`临时文件已删除: ${tempPdfPath}`);
} catch (cleanupError) {
console.warn(`清理临时文件失败: ${cleanupError}`);
}
}
}
}
// 注册工具列表处理器
server.setRequestHandler(ListToolsRequestSchema, async () => {
return {
tools: [
{
name: "search_arxiv",
description: "搜索 arXiv 论文",
inputSchema: {
type: "object",
properties: {
query: {
type: "string",
description: "搜索英文关键词"
},
maxResults: {
type: "number",
description: "最大结果数量",
default: 5
}
},
required: ["query"]
}
},
{
name: "get_recent_ai_papers",
description: "获取 arXiv AI 领域最新论文（cs.AI/recent）",
inputSchema: {
type: "object",
properties: {},
required: []
}
},
{
name: "get_arxiv_pdf_url",
description: "获取 arXiv PDF 下载链接",
inputSchema: {
type: "object",
properties: {
input: {
type: "string",
description: "arXiv 论文URL（如：http://arxiv.org/abs/2403.15137v1）或 arXiv ID（如：2403.15137v1）"
}
},
required: ["input"]
}
},
{
name: "parse_paper_content",
description: "解析论文内容（优先使用 HTML 版本，回退到 PDF）",
inputSchema: {
type: "object",
properties: {
input: {
type: "string",
description: "arXiv 论文URL或 arXiv ID"
},
paperInfo: {
type: "object",
description: "论文信息（可选，用于添加论文元数据）",
properties: {
title: { type: "string" },
summary: { type: "string" },
published: { type: "string" },
authors: { type: "array" }
}
}
},
required: ["input"]
}
}
]
};
});
// 注册工具调用处理器
server.setRequestHandler(CallToolRequestSchema, async (request) => {
const { name, arguments: args } = request.params;
try {
switch (name) {
case "search_arxiv": {
const { query, maxResults = 5 } = args as { query: string; maxResults?: number };
const results = await searchArxivPapers(query, maxResults);
return {
content: [{
type: "text",
text: `找到 ${results.papers.length} 篇相关论文（总计 ${results.totalResults} 篇）：\n\n${results.papers.map((paper, index) =>
`${index + 1}. **${paper.title}**\n ID: ${paper.id}\n 发布日期: ${paper.published}\n 作者: ${paper.authors.map((author: any) => author.name || author).join(', ')}\n 摘要: ${paper.summary.substring(0, 300)}...\n URL: ${paper.url}\n`
).join('\n')}`
}]
};
}
case "get_recent_ai_papers": {
const htmlContent = await getRecentAIPapers();
return {
content: [{
type: "text",
text: htmlContent
}]
};
}
case "get_arxiv_pdf_url": {
const { input } = args as { input: string };
const pdfUrl = getArxivPdfUrl(input);
return {
content: [{
type: "text",
text: `PDF 下载链接: ${pdfUrl}`
}]
};
}
case "parse_paper_content": {
const { input, paperInfo } = args as { input: string; paperInfo?: any };
const result = await parsePaperContent(input, paperInfo);
return {
content: [{
type: "text",
text: result.content
}]
};
}
default:
throw new Error(`Unknown tool: ${name}`);
}
} catch (error) {
return {
content: [{
type: "text",
text: `工具执行失败: ${error instanceof Error ? error.message : String(error)}`
}],
isError: true
};
}
});
// 启动服务器
console.log("启动 ArXiv Paper MCP Server...");
const transport = new StdioServerTransport();
await server.connect(transport);
console.log("🚀 ArXiv Paper MCP Server 已启动，等待连接...");