Skip to content
Back to search

paper-search-mcp-nodejs

A Node.js implementation of the Model Context Protocol (MCP) server for searching and downloading academic papers from multiple sources, including **Web of Science**, arXiv, and more.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/dianel555--paper-search-mcp-nodejs/badge.svg)](/listings/dianel555--paper-search-mcp-nodejs)

HTML

<a href="/listings/dianel555--paper-search-mcp-nodejs"><img src="/api/listings/dianel555--paper-search-mcp-nodejs/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

System Requirements

- Node.js 20.18.1+ (Node.js 21 is excluded by dependency engine constraints) - npm or yarn

Installation

# Clone repository
git clone https://github.com/your-username/paper-search-mcp-nodejs.git
cd paper-search-mcp-nodejs

# Install dependencies
npm install

# Copy environment template
cp .env.example .env

Configuration

1. **Get Web of Science API Key** - Visit [Clarivate Developer Portal](https://developer.clarivate.com/apis) - Register and apply for Web of Science API access - Add API key to `.env` file

2. **Get PubMed API Key (Optional)** - Without API key: Free usage, 3 requests/second limit - With API key: 10 requests/second, more stable service - Get key: See [NCBI API Keys](https://ncbiinsights.ncbi.nlm.nih.gov/2017/11/02/new-api-keys-for-the-e-utilities/)

3. **Configure Environment Variables**

   # Edit .env file
   # Web of Science defaults to Starter v2 and uses WOS_API_KEY.
   WOS_API_KEY=your_web_of_science_api_key
   # Optional Expanded product key.
   WOS_EXPANDED_API_KEY=your_expanded_key
   WOS_STARTER_VERSION=v2
   WOS_STARTER_RPS=1
   WOS_STARTER_DAILY_LIMIT=50
   WOS_EXPANDED_RPS=2
   # Full Record records/day; 0 means unlimited local accounting.
   WOS_EXPANDED_FULL_RECORD_BUDGET=0
   WOS_EXPANDED_BASE_URL=https://api.clarivate.com/api/wos

   # Optional public-page HTML fallback; never a WoS/API proxy.
   # A key alone does not authorize paid retrieval; browser and residential
   # escalation are separate opt-ins. Invalid/missing paid configuration keeps Direct available.
   # Without residential authorization Publisher/Scholar default to 50 credits/operation and 10/request.
   # With SCRAPINGANT_ALLOW_RESIDENTIAL=true their defaults become 500/125;
   # explicit limits always win and a residential request still requires the residential ceiling.
   SCRAPINGANT_API_KEY=
   SCRAPINGANT_ENABLED=false
   SCRAPINGANT_ALLOW_BROWSER_ESCALATION=false
   SCRAPINGANT_ALLOW_RESIDENTIAL=false
   SCRAPINGANT_MAX_CREDITS_PER_OPERATION=50
   SCRAPINGANT_MAX_CREDITS_PER_REQUEST=10
   SCRAPINGANT_MAX_CONCURRENCY=1
   SCRAPINGANT_PROXY_TYPE=datacenter

   # Controlled Sci-Hub adapter; disabled unless explicitly enabled.
   # Mirror addresses are discovered from the Sci-Hub and ooopn directory pages.
   # SCIHUB_MIRRORS is optional and accepts comma-separated supplemental URLs.
   SCIHUB_ENABLED=false
   SCIHUB_FETCH_MODE=fallback
   SCIHUB_MIRRORS=
   SCIHUB_HEALTHCHECK_CONCURRENCY=3

   # PubMed API key (optional, recommended for better performance)
   PUBMED_API_KEY=your_ncbi_api_key_here

   # Semantic Scholar API key (optional, increases rate limits)
   SEMANTIC_SCHOLAR_API_KEY=your_semantic_scholar_api_key

   # Elsevier API key: ScienceDirect Search v2 and Scopus details
   ELSEVIER_API_KEY=your_elsevier_api_key
   # Optional dedicated key for Scopus Search API; falls back to ELSEVIER_API_KEY
   SCOPUS_SEARCH_API_KEY=

   # Springer Nature API keys (required for Springer)
   SPRINGER_API_KEY=your_springer_api_key  # For Metadata API v2
   # Optional: Separate key for OpenAccess API (if different from main key)
   SPRINGER_OPENACCESS_API_KEY=your_openaccess_api_key

   # Wiley TDM token (required for Wiley)
   WILEY_TDM_TOKEN=your_wiley_tdm_token

Build and Run

Method 1: NPX (Recommended for MCP)

# Direct run with npx (most common MCP deployment)
npx -y paper-search-mcp-nodejs

# Or install globally
npm install -g paper-search-mcp-nodejs
paper-search-mcp

Method 2: Local Development

# Build TypeScript code
npm run build

# Start server
npm start

# Or run in development mode
npm run dev

MCP Server Configuration

Add the following configuration to your Claude Desktop config file:

**macOS**: `~/Library/Application Support/Claude/claude_desktop_config.json` **Windows**: `%APPDATA%\Claude\claude_desktop_config.json`

Minimal NPX Configuration (Start Here)

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

README.md · README · 882 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. # Paper Search MCP (Node.js)
  2. ## English|[中文](README-sc.md)
  3. A Node.js Model Context Protocol (MCP) server for searching and downloading academic papers from multiple sources, including arXiv, Web of Science, PubMed, Google Scholar, Sci-Hub, ScienceDirect, Springer, Wiley, Scopus, Crossref, and **14 academic platforms** in total.
  4. ![Node.js](https://img.shields.io/badge/node.js-20.18.1%2B%20%2F%2022%2B-green.svg)
  5. ![TypeScript](https://img.shields.io/badge/typescript-^5.5.3-blue.svg)
  6. ![License](https://img.shields.io/badge/license-MIT-blue.svg)
  7. ![Platforms](https://img.shields.io/badge/platforms-14-brightgreen.svg)
  8. ![Version](https://img.shields.io/badge/version-0.3.3-blue.svg)
  9. ## 💖 Sponsors
  10. <p>
  11. <a href="https://scrapingant.com/">
  12. <img src="assets/scrapingant.png" alt="ScrapingAnt" width="60" height="60">
  13. </a>
  14. </p>
  15. <a href="https://scrapingant.com/">
  16. <img src="assets/scrapingant-banner.png" alt="ScrapingAnt web scraping service" width="640">
  17. </a>
  18. This project is sponsored by [ScrapingAnt](https://scrapingant.com/), a web scraping service for accessing public web data.
  19. **Offer for project users:** Use code `ENTHUSIAST_50` for **50% off the first month of the Enthusiast plan**. The discount applies to the first month only.
  20. ## ✨ Key Features
  21. - **🌍 14 Academic Platforms**: arXiv, Web of Science, PubMed, Google Scholar, bioRxiv, medRxiv, Semantic Scholar, IACR ePrint, Sci-Hub, ScienceDirect, Springer Nature, Wiley, Scopus, Crossref
  22. - **🧭 WoS Starter + Expanded**: Starter v2 by default; Expanded SR/FR and citation relationships only when explicitly selected
  23. - **🌐 Public-page access discovery**: optional ScrapingAnt Extended fetching for DOI publisher pages; never used for Clarivate API or institutional login pages
  24. - **🔗 MCP Protocol Integration**: Seamless integration with Claude Desktop and other AI assistants
  25. - **📊 Unified Data Model**: Standardized paper format across all platforms
  26. - **⚡ High-Performance Search**: Concurrent search with intelligent rate limiting
  27. - **🛡️ Security First**: DOI validation, query sanitization, injection prevention, sensitive data masking
  28. - **📝 Type Safety**: Complete TypeScript support with extended interfaces
  29. - **🎯 Academic Papers First**: Smart filtering prioritizing academic papers over books
  30. - **🔄 Smart Error Handling**: Unified ErrorHandler with retry logic and platform fallback
  31. ## 📚 Supported Platforms
  32. | Platform | Search | Download | Full Text | Citations | API Key | Special Features |
  33. |----------|--------|----------|-----------|-----------|---------|------------------|
  34. | **Crossref** | ✅ | ❌ | ❌ | ✅ | ❌ | Default search, extensive metadata coverage |
  35. | **arXiv** | ✅ | ✅ | ✅ | ❌ | ❌ | Physics/CS preprints |
  36. | **Web of Science** | ✅ | ❌ | ❌ | ✅ | ✅ Required | Starter v2 default; Expanded SR/FR and relations opt-in |
  37. | **PubMed** | ✅ | ❌ | ❌ | ❌ | 🟡 Optional | Biomedical literature |
  38. | **Google Scholar** | ✅ | ❌ | ❌ | ✅ | ❌ | Direct parser or optional ScrapingAnt General endpoint |
  39. | **bioRxiv** | ✅ | ✅ | ✅ | ❌ | ❌ | Biology preprints |
  40. | **medRxiv** | ✅ | ✅ | ✅ | ❌ | ❌ | Medical preprints |
  41. | **Semantic Scholar** | ✅ | ✅ | ❌ | ✅ | 🟡 Optional | AI semantic search |
  42. | **IACR ePrint** | ✅ | ✅ | ✅ | ❌ | ❌ | Cryptography papers |
  43. | **Sci-Hub** | Opt-in | Opt-in | ❌ | ❌ | ❌ | Controlled DOI-only HTML adapter; disabled by default |
  44. | **ScienceDirect** | ✅ | ❌ | ❌ | ✅ | ✅ Required | Elsevier's full-text database |
  45. | **Springer Nature** | ✅ | ✅* | ❌ | ❌ | ✅ Required | Dual API: Meta v2 & OpenAccess |
  46. | **Wiley** | ❌ | ✅ | ✅ | ❌ | ✅ Required | TDM API: DOI-based PDF download only |
  47. | **Scopus** | ✅ | ❌ | ❌ | ✅ | ✅ Required | Largest citation database |
  48. ✅ Supported | ❌ Not supported | 🟡 Optional | ✅* Open Access only
  49. > **Note**: Wiley TDM API does not support keyword search. Use `search_crossref` to find Wiley articles, then use `download_paper` with `platform="wiley"` to download PDFs by DOI.
  50. ## ⚖️ Compliance & Ethical Use (Sci-Hub / Google Scholar)
  51. This project includes integrations that may have **legal, contractual (ToS), and ethical** constraints. You are responsible for ensuring your usage complies with applicable laws, institutional policies, and third‑party terms.
  52. - **Sci-Hub**: Disabled by default and limited to an unstable DOI/mirror adapter. It does not grant access rights; enable it only for content you are legally authorized to access.
  53. - **Google Scholar/ScrapingAnt**: Automated fetching may trigger blocking or contractual restrictions. ScrapingAnt is used only for public Scholar or publisher pages when configured, never for WoS login, SSO, MFA, or institutional subscription pages.
  54. ## 🚀 Quick Start
  55. ### System Requirements
  56. - Node.js 20.18.1+ (Node.js 21 is excluded by dependency engine constraints)
  57. - npm or yarn
  58. ### Installation
  59. ```bash
  60. # Clone repository
  61. git clone https://github.com/your-username/paper-search-mcp-nodejs.git
  62. cd paper-search-mcp-nodejs
  63. # Install dependencies
  64. npm install
  65. # Copy environment template
  66. cp .env.example .env
  67. ```
  68. ### Configuration
  69. 1. **Get Web of Science API Key**
  70. - Visit [Clarivate Developer Portal](https://developer.clarivate.com/apis)
  71. - Register and apply for Web of Science API access
  72. - Add API key to `.env` file
  73. 2. **Get PubMed API Key (Optional)**
  74. - Without API key: Free usage, 3 requests/second limit
  75. - With API key: 10 requests/second, more stable service
  76. - Get key: See [NCBI API Keys](https://ncbiinsights.ncbi.nlm.nih.gov/2017/11/02/new-api-keys-for-the-e-utilities/)
  77. 3. **Configure Environment Variables**
  78. ```bash
  79. # Edit .env file
  80. # Web of Science defaults to Starter v2 and uses WOS_API_KEY.
  81. WOS_API_KEY=your_web_of_science_api_key
  82. # Optional Expanded product key.
  83. WOS_EXPANDED_API_KEY=your_expanded_key
  84. WOS_STARTER_VERSION=v2
  85. WOS_STARTER_RPS=1
  86. WOS_STARTER_DAILY_LIMIT=50
  87. WOS_EXPANDED_RPS=2
  88. # Full Record records/day; 0 means unlimited local accounting.
  89. WOS_EXPANDED_FULL_RECORD_BUDGET=0
  90. WOS_EXPANDED_BASE_URL=https://api.clarivate.com/api/wos
  91. # Optional public-page HTML fallback; never a WoS/API proxy.
  92. # A key alone does not authorize paid retrieval; browser and residential
  93. # escalation are separate opt-ins. Invalid/missing paid configuration keeps Direct available.
  94. # Without residential authorization Publisher/Scholar default to 50 credits/operation and 10/request.
  95. # With SCRAPINGANT_ALLOW_RESIDENTIAL=true their defaults become 500/125;
  96. # explicit limits always win and a residential request still requires the residential ceiling.
  97. SCRAPINGANT_API_KEY=
  98. SCRAPINGANT_ENABLED=false
  99. SCRAPINGANT_ALLOW_BROWSER_ESCALATION=false
  100. SCRAPINGANT_ALLOW_RESIDENTIAL=false
  101. SCRAPINGANT_MAX_CREDITS_PER_OPERATION=50
  102. SCRAPINGANT_MAX_CREDITS_PER_REQUEST=10
  103. SCRAPINGANT_MAX_CONCURRENCY=1
  104. SCRAPINGANT_PROXY_TYPE=datacenter
  105. # Controlled Sci-Hub adapter; disabled unless explicitly enabled.
  106. # Mirror addresses are discovered from the Sci-Hub and ooopn directory pages.
  107. # SCIHUB_MIRRORS is optional and accepts comma-separated supplemental URLs.
  108. SCIHUB_ENABLED=false
  109. SCIHUB_FETCH_MODE=fallback
  110. SCIHUB_MIRRORS=
  111. SCIHUB_HEALTHCHECK_CONCURRENCY=3
  112. # PubMed API key (optional, recommended for better performance)
  113. PUBMED_API_KEY=your_ncbi_api_key_here
  114. # Semantic Scholar API key (optional, increases rate limits)
  115. SEMANTIC_SCHOLAR_API_KEY=your_semantic_scholar_api_key
  116. # Elsevier API key: ScienceDirect Search v2 and Scopus details
  117. ELSEVIER_API_KEY=your_elsevier_api_key
  118. # Optional dedicated key for Scopus Search API; falls back to ELSEVIER_API_KEY
  119. SCOPUS_SEARCH_API_KEY=
  120. # Springer Nature API keys (required for Springer)
  121. SPRINGER_API_KEY=your_springer_api_key # For Metadata API v2
  122. # Optional: Separate key for OpenAccess API (if different from main key)
  123. SPRINGER_OPENACCESS_API_KEY=your_openaccess_api_key
  124. # Wiley TDM token (required for Wiley)
  125. WILEY_TDM_TOKEN=your_wiley_tdm_token
  126. ```
  127. ### Build and Run
  128. #### Method 1: NPX (Recommended for MCP)
  129. ```bash
  130. # Direct run with npx (most common MCP deployment)
  131. npx -y paper-search-mcp-nodejs
  132. # Or install globally
  133. npm install -g paper-search-mcp-nodejs
  134. paper-search-mcp
  135. ```
  136. #### Method 2: Local Development
  137. ```bash
  138. # Build TypeScript code
  139. npm run build
  140. # Start server
  141. npm start
  142. # Or run in development mode
  143. npm run dev
  144. ```
  145. ### MCP Server Configuration
  146. Add the following configuration to your Claude Desktop config file:
  147. **macOS**: `~/Library/Application Support/Claude/claude_desktop_config.json`
  148. **Windows**: `%APPDATA%\Claude\claude_desktop_config.json`
  149. #### Minimal NPX Configuration (Start Here)
  150. Use this minimal configuration to start immediately with public sources such as Crossref and arXiv. No API key or optional environment variable is required; add platform-specific keys only when you need them.
  151. ```json
  152. {
  153. "mcpServers": {
  154. "paper-search-nodejs": {
  155. "command": "npx",
  156. "args": ["-y", "paper-search-mcp-nodejs"]
  157. }
  158. }
  159. }
  160. ```
  161. #### Complete NPX Configuration (Advanced/Optional)
  162. Use the complete configuration below only when you need keyed platforms, optional fallbacks, or custom limits. Every value in `env` must be a string; replace placeholders with values from your own environment, and never commit real credentials.
  163. ```json
  164. {
  165. "mcpServers": {
  166. "paper-search-nodejs": {
  167. "command": "npx",
  168. "args": ["-y", "paper-search-mcp-nodejs"],
  169. "env": {
  170. "NODE_ENV": "production",
  171. "LOG_LEVEL": "info",
  172. "WOS_API_KEY": "your_web_of_science_api_key",
  173. "WOS_STARTER_VERSION": "v2",
  174. "WOS_STARTER_RPS": "1",
  175. "WOS_STARTER_DAILY_LIMIT": "50",
  176. "WOS_EXPANDED_API_KEY": "",
  177. "WOS_EXPANDED_RPS": "2",
  178. "WOS_EXPANDED_FULL_RECORD_BUDGET": "0",
  179. "WOS_EXPANDED_BASE_URL": "https://api.clarivate.com/api/wos",
  180. "PUBMED_API_KEY": "",
  181. "SEMANTIC_SCHOLAR_API_KEY": "",
  182. "ELSEVIER_API_KEY": "",
  183. "SCOPUS_SEARCH_API_KEY": "",
  184. "SPRINGER_API_KEY": "",
  185. "SPRINGER_OPENACCESS_API_KEY": "",
  186. "WILEY_TDM_TOKEN": "",
  187. "CROSSREF_MAILTO": "you@example.com",
  188. "SCRAPINGANT_API_KEY": "",
  189. "SCRAPINGANT_ENABLED": "false",
  190. "SCRAPINGANT_ALLOW_BROWSER_ESCALATION": "false",
  191. "SCRAPINGANT_ALLOW_RESIDENTIAL": "false",
  192. "SCRAPINGANT_MAX_CREDITS_PER_OPERATION": "50",
  193. "SCRAPINGANT_MAX_CREDITS_PER_REQUEST": "10",
  194. "SCRAPINGANT_MAX_CONCURRENCY": "1",
  195. "SCRAPINGANT_PROXY_TYPE": "datacenter",
  196. "SCHOLAR_PROXY": "",
  197. "SCIHUB_ENABLED": "false",
  198. "SCIHUB_FETCH_MODE": "fallback",
  199. "SCIHUB_MIRRORS": "",
  200. "SCIHUB_HEALTHCHECK_CONCURRENCY": "3",
  201. "DEFAULT_DOWNLOAD_PATH": "./downloads",
  202. "MAX_FILE_SIZE_MB": "100",
  203. "RATE_LIMIT_REQUESTS_PER_MINUTE": "60",
  204. "RATE_LIMIT_BURST": "10"
  205. }
  206. }
  207. }
  208. }
  209. ```
  210. `SCHOLAR_PROXY` is an optional explicit override for Scholar direct transport. When it is empty, Scholar uses the standard local proxy aliases (`HTTPS_PROXY`/`HTTP_PROXY`/`ALL_PROXY`, including lowercase forms) when configured; when set, it takes precedence and is parsed as a full HTTP(S)/SOCKS proxy URL. It does not select the ScrapingAnt paid fallback. The retrieval benchmark does not enable or replace it. `WOS_EXPANDED_API_KEY`, `SCRAPINGANT_API_KEY`, and the other optional keys may remain empty.
  211. #### Local Installation Configuration
  212. For a local build, keep the complete `env` object above and change only the server command:
  213. ```json
  214. {
  215. "command": "node",
  216. "args": ["/path/to/paper-search-mcp-nodejs/dist/server.js"]
  217. }
  218. ```
  219. ## 🛠️ MCP Tools
  220. ### `search_papers`
  221. Search academic papers across multiple platforms
  222. ```typescript
  223. // Random platform selection (default behavior)
  224. search_papers({
  225. query: "machine learning",
  226. platform: "all", // Randomly selects one platform for efficiency
  227. maxResults: 10,
  228. year: "2023",
  229. sortBy: "date"
  230. })
  231. // Search specific platform
  232. search_papers({
  233. query: "quantum computing",
  234. platform: "webofscience", // Target specific platform
  235. maxResults: 5
  236. })
  237. ```
  238. **Platform Selection Behavior:**
  239. - `platform: "crossref"` (default) - Free API with extensive scholarly metadata coverage
  240. - `platform: "all"` - Randomly selects one platform for efficient, focused results
  241. - Specific platform - Searches only that platform
  242. - Available platforms: `crossref`, `arxiv`, `webofscience`/`wos`, `pubmed`, `biorxiv`, `medrxiv`, `semantic`, `iacr`, `googlescholar`/`scholar`, `scihub`, `sciencedirect`, `springer`, `scopus`
  243. - Note: `wiley` only supports PDF download by DOI, not keyword search
  244. ### `search_crossref`
  245. Search academic papers from Crossref database (default search platform)
  246. ```typescript
  247. search_crossref({
  248. query: "machine learning",
  249. maxResults: 10,
  250. year: "2023",
  251. author: "Smith",
  252. sortBy: "relevance", // or "date", "citations"
  253. sortOrder: "desc"
  254. })
  255. ```
  256. ### `search_arxiv`
  257. Search arXiv preprints specifically
  258. ```typescript
  259. search_arxiv({
  260. query: "transformer neural networks",
  261. maxResults: 10,
  262. category: "cs.AI",
  263. author: "Vaswani",
  264. year: "2023",
  265. sortBy: "date", // relevance, date, citations
  266. sortOrder: "desc" // asc, desc
  267. })
  268. ```
  269. ### `search_webofscience`
  270. Search Web of Science database specifically
  271. ```typescript
  272. search_webofscience({
  273. query: "CRISPR gene editing",
  274. maxResults: 5,
  275. year: "2022",
  276. journal: "Nature",
  277. apiProduct: "expanded", // omit for Starter v2
  278. recordView: "short", // expanded only; "full" is opt-in
  279. discoverAccess: true, // optional publisher public-page discovery
  280. discoverAccessMaxItems: 5 // explicit bound; default is 5, range 1-100
  281. })
  282. get_webofscience_related_records({
  283. uid: "WOS:000000000000001",
  284. relation: "citing", // references, citing, or related
  285. maxResults: 50
  286. })
  287. ```
  288. ### `search_pubmed`
  289. Search PubMed/MEDLINE biomedical literature database
  290. ```typescript
  291. search_pubmed({
  292. query: "COVID-19 vaccine efficacy",
  293. maxResults: 20,
  294. year: "2023",
  295. author: "Smith",
  296. journal: "New England Journal of Medicine",
  297. publicationType: ["Journal Article", "Clinical Trial"],
  298. sortBy: "date" // relevance, date
  299. })
  300. ```
  301. ### `search_google_scholar`
  302. Search Google Scholar academic database
  303. ```typescript
  304. search_google_scholar({
  305. query: "machine learning",
  306. maxResults: 10,
  307. yearLow: 2020,
  308. yearHigh: 2023,
  309. author: "Bengio"
  310. })
  311. ```
  312. ### `search_biorxiv` / `search_medrxiv`
  313. Search biology and medical preprints
  314. ```typescript
  315. search_biorxiv({
  316. query: "CRISPR",
  317. maxResults: 15,
  318. days: 30,
  319. category: "genomics" // neuroscience, genomics, etc.
  320. })
  321. search_medrxiv({
  322. query: "COVID-19",
  323. maxResults: 10,
  324. days: 30,
  325. category: "infectious_diseases"
  326. })
  327. ```
  328. ### `search_semantic_scholar`
  329. Search Semantic Scholar AI semantic database
  330. ```typescript
  331. search_semantic_scholar({
  332. query: "deep learning",
  333. maxResults: 10,
  334. fieldsOfStudy: ["Computer Science"],
  335. year: "2023"
  336. })
  337. ```
  338. ### `search_iacr`
  339. Search IACR ePrint cryptography archive
  340. ```typescript
  341. search_iacr({
  342. query: "zero knowledge proof",
  343. maxResults: 5,
  344. fetchDetails: true
  345. })
  346. ```
  347. ### `search_scihub`
  348. Controlled, opt-in Sci-Hub DOI lookup/download; disabled by default and not an official API
  349. ```typescript
  350. // Requires SCIHUB_ENABLED=true. DOI/doi.org inputs only.
  351. // Mirrors are discovered from https://sci-hub.mobi/en/mirrors and
  352. // https://www.ooopn.com/tool/scihub/; SCIHUB_MIRRORS adds optional comma-separated URLs.
  353. search_scihub({
  354. doiOrUrl: "10.1038/nature12373",
  355. downloadPdf: true,
  356. savePath: "./downloads"
  357. })
  358. ```
  359. ### `search_sciencedirect`
  360. Search Elsevier ScienceDirect database
  361. ```typescript
  362. search_sciencedirect({
  363. query: "artificial intelligence",
  364. maxResults: 10,
  365. year: "2023",
  366. author: "Smith",
  367. openAccess: true // Filter for open access articles
  368. })
  369. ```
  370. ### `search_springer`
  371. Search Springer Nature database (Metadata API v2 or OpenAccess API)
  372. ```typescript
  373. search_springer({
  374. query: "machine learning",
  375. maxResults: 10,
  376. year: "2023",
  377. openAccess: true, // Use OpenAccess API for downloadable PDFs
  378. type: "Journal" // Filter: Journal, Book, or Chapter
  379. })
  380. ```
  381. ### `search_scopus`
  382. Search Scopus citation database
  383. ```typescript
  384. search_scopus({
  385. query: "renewable energy",
  386. maxResults: 10,
  387. year: "2023",
  388. affiliation: "MIT",
  389. documentType: "ar" // ar=article, cp=conference, re=review
  390. })
  391. ```
  392. Scopus search requests `COMPLETE` by default. If Elsevier explicitly denies the `COMPLETE` view because the key lacks that entitlement, the search enters one bounded `STANDARD` fallback strategy; any transient retries remain subject to the existing retry policy. The fallback omits the `field` override so the API can return its standard field set. Other authentication, query, rate-limit, network, and server errors are not converted into a view fallback. `STANDARD` may contain less enriched metadata (for example, full author, abstract, keyword, affiliation, and funding fields); a valid Scopus API key is still required. See [Scopus Search API views](https://dev.elsevier.com/sc_search_views.html) for the fields available in each view.
  393. ### `check_scihub_mirrors`
  394. Check health status of Sci-Hub mirror sites
  395. ```typescript
  396. check_scihub_mirrors({
  397. forceCheck: true // Force fresh health check
  398. })
  399. ```
  400. ### `download_paper`
  401. Download paper PDF files
  402. ```typescript
  403. download_paper({
  404. paperId: "2106.12345", // or DOI for Sci-Hub
  405. platform: "arxiv", // or "scihub" for Sci-Hub downloads
  406. savePath: "./downloads"
  407. })
  408. ```
  409. ### `get_paper_by_doi`
  410. Get paper information by DOI
  411. ```typescript
  412. get_paper_by_doi({
  413. doi: "10.1038/s41586-023-12345-6",
  414. platform: "all"
  415. })
  416. ```
  417. ### `discover_paper_access`
  418. Discover one public publisher PDF candidate by DOI. This does not claim open-license status or complete download success; PDF verification is opt-in and bounded.
  419. ```typescript
  420. discover_paper_access({
  421. doi: "https://doi.org/10.1038/s41586-023-12345-6",
  422. verifyPdf: false
  423. })
  424. ```
  425. ### `get_platform_status`
  426. Check local platform capability and API-key status. This is a local diagnostic and does not validate ScrapingAnt account quota or make a paid request.
  427. ```typescript
  428. get_platform_status({})
  429. ```
  430. ### Public access and cost boundaries
  431. `discover_paper_access` accepts only a DOI. It returns a bounded access state such as `oa_candidate`, `pdf_verified`, `not_found`, `restricted`, `failed`, or `skipped`; a candidate is not an open-license or complete-download claim. `verifyPdf` is opt-in and performs only a bounded PDF prefix check. Candidate order, source provenance, HTTP/API status, fallback attempts, and known local cost are separate evidence fields.
  432. Paid retrieval is Direct-first and finite. Empty/parse-failed public pages may consume an explicitly authorized fallback attempt; a known permission, unsafe target, resource limit, cancellation, deadline, unknown price, or closed ledger stops paid work. Missing/invalid post-dispatch billing is recorded as unknown and consumes its local estimate, but does not independently stop a bounded retry/fallback chain; the final reported credits remain unknown until reconciled. When browser escalation is explicitly enabled, Scholar production tries ScrapingAnt `browser:datacenter` before the remaining paid combinations; browser is one bounded dispatch and has no retry. Returned diagnostics never include cookies, authorization, queries, raw HTML, or sensitive URLs. Public cookies are not read from configuration; Scholar session cookies, when obtained, stay on the exact Scholar HTTPS origin and are never sent to ScrapingAnt.
  433. Use `SCRAPINGANT_ENABLED=true` only after obtaining deployment authorization. Scholar's browser-first fallback requires `SCRAPINGANT_ALLOW_BROWSER_ESCALATION=true`; `SCRAPINGANT_ALLOW_BROWSER_ESCALATION` and `SCRAPINGANT_ALLOW_RESIDENTIAL` are independent controls. Restarting with either flag disabled rolls back the corresponding combinations; it does not erase already observed local costs. A run can also be disabled by leaving the key/paid flag off. No setting synchronizes provider quota or clears an in-flight ledger.
  434. ### Offline benchmark
  435. The fixed benchmark corpus and strategy schedule are validated without network access. A complete 360-cell live run is a separately authorized external qualification, not the completion gate for each engineering fix:
  436. ```bash
  437. # Accounting/report simulation (no production retrieval workflow)
  438. npm run --silent benchmark:offline -- --json
  439. # Reviewed offline production workflow beneath fixed raw fixtures (virtual clock)
  440. npm run --silent benchmark:offline -- --workflow --json
  441. # Optional: write encoded run-id .json and .md artifacts without overwriting existing files.
  442. npm run --silent benchmark:offline -- --workflow --output-dir ./benchmark-artifacts
  443. ```
  444. The default command validates the frozen 20 DOI/10 Scholar query corpus with an injected response-only accounting simulator. Add `--workflow` to run the reviewed `createProductionBenchmarkCellExecutor` instead: it invokes the real Publisher/Scholar business entries, providers, parsing, session, fallback, scheduler, billing bridge and PDF-prefix paths beneath independent fixed raw fixtures, using a virtual clock while preserving the production pacing rules. Both modes are explicitly offline-only; neither initializes a live provider or uses credentials for retrieval, and neither can produce `live_passed`. The workflow report must be identified separately from simulator totals. Live evaluation is exposed only through the separately named command below; it requires explicit `--authorize-live`, an exclusive explicit `--run-id`, a complete preflight, and the fixed safety ceiling of 15,000 credits, 1,500 HTTP dispatches, and 7,200,000ms. A failed preflight writes a privacy-safe `blocked` report with zero dispatches; it never shrinks the frozen matrix or bypasses configuration. The live executor uses real production transports and is not the offline fixture executor. The planned schedule is 60 production cells plus 300 independent comparison cells; unexecuted live cells remain `not_run` and keep their denominator. Artifact paths are exclusive so a report cannot silently overwrite an earlier run.
  445. ```bash
  446. # Explicitly authorized live campaign; preflight blocks without complete paid/browser/residential configuration.
  447. npm run --silent benchmark:live -- --authorize-live --authorize-scholar-proxy --run-id live-YYYYMMDD-01 --output-dir ./live-benchmark-artifacts --json
  448. ```
  449. Do not pass `--authorize-live` unless the campaign, provider billing, target scope, and live-capable harness have been reviewed. Pass `--authorize-scholar-proxy` only when the explicit `SCHOLAR_PROXY` endpoint has separately passed TLS/ownership review; ambient proxy aliases alone remain blocked. The command never resumes an old run, appends budget, or changes `offlineOnly` fixtures.
  450. ## 📊 Data Model
  451. All platform paper data is converted to a unified format:
  452. ```typescript
  453. interface Paper {
  454. paperId: string; // Unique identifier
  455. title: string; // Paper title
  456. authors: string[]; // Author list
  457. abstract: string; // Abstract
  458. doi: string; // DOI
  459. publishedDate: Date; // Publication date
  460. pdfUrl: string; // PDF link
  461. url: string; // Paper page URL
  462. source: string; // Source platform
  463. citationCount?: number; // Citation count
  464. journal?: string; // Journal name
  465. year?: number; // Publication year
  466. categories?: string[]; // Subject categories
  467. keywords?: string[]; // Keywords
  468. // ... more fields
  469. }
  470. ```
  471. ## 🔧 Development
  472. ### Project Structure
  473. ```
  474. src/
  475. ├── models/
  476. │ └── Paper.ts # Paper data model
  477. ├── platforms/
  478. │ ├── PaperSource.ts # Abstract base class
  479. │ ├── ArxivSearcher.ts # arXiv searcher
  480. │ ├── WebOfScienceSearcher.ts # Web of Science searcher
  481. │ ├── PubMedSearcher.ts # PubMed searcher
  482. │ ├── GoogleScholarSearcher.ts # Google Scholar searcher
  483. │ ├── BioRxivSearcher.ts # bioRxiv/medRxiv searcher
  484. │ ├── SemanticScholarSearcher.ts # Semantic Scholar searcher
  485. │ ├── IACRSearcher.ts # IACR ePrint searcher
  486. │ ├── SciHubSearcher.ts # Sci-Hub searcher with mirror management
  487. │ ├── ScienceDirectSearcher.ts # ScienceDirect (Elsevier) searcher
  488. │ ├── SpringerSearcher.ts # Springer Nature searcher (Meta v2 & OpenAccess APIs)
  489. │ ├── WileySearcher.ts # Wiley TDM API (DOI-based PDF download only)
  490. │ ├── ScopusSearcher.ts # Scopus citation database searcher
  491. │ └── CrossrefSearcher.ts # Crossref API searcher (default platform)
  492. ├── mcp/
  493. │ ├── tools.ts # MCP tool definitions
  494. │ ├── schemas.ts # Zod schemas for tool arguments
  495. │ ├── handleToolCall.ts # Tool call dispatcher
  496. │ └── searchers.ts # Searcher initialization
  497. ├── utils/
  498. │ ├── SecurityUtils.ts # DOI validation, query sanitization, injection prevention
  499. │ ├── PublicNetwork.ts # Public-target DNS/redirect and SSRF checks
  500. │ ├── ConcurrencyLimiter.ts # Dependency-free bounded concurrency
  501. │ ├── ErrorHandler.ts # Unified error handling with retry logic
  502. │ ├── RateLimiter.ts # Token bucket rate limiting
  503. │ ├── QuotaManager.ts # Daily quota tracking
  504. │ ├── RequestCache.ts # LRU caching for requests
  505. │ ├── PDFExtractor.ts # PDF text extraction
  506. │ └── Logger.ts # Debug logging
  507. ├── config/
  508. │ └── constants.ts # Timeouts, endpoints, limits
  509. ├── services/
  510. │ ├── CitationService.ts # Citation fetching service
  511. │ ├── WebOfScienceParser.ts # Starter/Expanded response parsers
  512. │ ├── WebOfScienceRequestService.ts # WoS retry, rate, quota, and status
  513. │ ├── PublicHttpClient.ts # Redirect-checked public HTTP
  514. │ ├── ScrapingAntFetcher.ts # General/Extended HTML API wrapper
  515. │ └── PublicAccessDiscovery.ts # DOI publisher-page PDF discovery
  516. └── server.ts # MCP server main file
  517. ```
  518. ### Adding New Platforms
  519. 1. Create new searcher class extending `PaperSource`
  520. 2. Implement required abstract methods
  521. 3. Register new searcher in `searchers.ts`
  522. 4. Add corresponding MCP tool in `tools.ts`
  523. ### Security Best Practices
  524. - All DOIs are validated before use in URLs
  525. - Query parameters are escaped to prevent injection
  526. - API keys are masked in all log output
  527. - Request timeouts prevent hanging connections
  528. - Query complexity limits prevent DoS attacks
  529. - Rate limiting and quota management prevent API abuse
  530. - Caching reduces external API calls
  531. ### Testing
  532. ```bash
  533. # Run tests
  534. npm test
  535. # Run linting
  536. npm run lint
  537. # Code formatting
  538. npm run format
  539. ```
  540. **Test Coverage:**
  541. - Includes WoS HTTP contracts, ScrapingAnt status/credits, public-target SSRF checks, access discovery, Sci-Hub fallback/PDF validation, and MCP schemas.
  542. - Platform searchers covered by unit and contract tests
  543. - Security utilities (DOI validation, query sanitization)
  544. - ErrorHandler (error classification, retry logic)
  545. - Rate limiting integration, QuotaManager, RequestCache
  546. | Test Suite | Coverage |
  547. |------------|----------|
  548. | Platform Searchers | ✅ |
  549. | SecurityUtils | ✅ |
  550. | ErrorHandler | ✅ |
  551. | RateLimiter & Integration | ✅ |
  552. | QuotaManager | ✅ |
  553. | RequestCache | ✅ |
  554. ## 🌟 Platform-Specific Features
  555. ### Springer Nature Dual API System
  556. Springer Nature provides two APIs:
  557. 1. **Metadata API v2** (Main API)
  558. - Endpoint: `https://api.springernature.com/meta/v2/json`
  559. - Searches all Springer content (subscription + open access)
  560. - Requires API key from https://dev.springernature.com/
  561. 2. **OpenAccess API** (Optional)
  562. - Endpoint: `https://api.springernature.com/openaccess/json`
  563. - Only searches open access content
  564. - May require separate API key or special permissions
  565. - Better for finding downloadable PDFs
  566. ```typescript
  567. // Search all Springer content
  568. search_springer({
  569. query: "machine learning",
  570. maxResults: 10
  571. })
  572. // Search only open access papers
  573. search_springer({
  574. query: "COVID-19",
  575. openAccess: true, // Uses OpenAccess API if available
  576. maxResults: 5
  577. })
  578. ```
  579. ### Web of Science Advanced Search
  580. 🎯 **WoS Starter + Expanded**: Starter API v2 is the default and v1 remains an explicit compatibility choice. Expanded must be selected explicitly.
  581. **API Version and product configuration:**
  582. ```bash
  583. # Starter version (default: v2; fixed for the process)
  584. WOS_STARTER_VERSION=v2
  585. # WOS_STARTER_VERSION=v1
  586. # Web of Science defaults to Starter v2.
  587. WOS_API_KEY=...
  588. # Optional Expanded product key.
  589. WOS_EXPANDED_API_KEY=...
  590. ```
  591. Starter requests use the documented `/documents` endpoints, page at most 50 records, and preserve unknown citation counts as `null`. Expanded uses its separate `/api/wos` contract, defaults to Short Record, and supports Full Record, references, citing, and related-record operations only when requested. The default is the current Swagger server `https://api.clarivate.com/api/wos`; older `wos-api.clarivate.com` guidance is not used.
  592. ```typescript
  593. // Multi-topic search
  594. search_webofscience({
  595. query: 'oriented structure',
  596. year: '2023-2025',
  597. sortBy: 'date',
  598. sortOrder: 'desc',
  599. maxResults: 10
  600. })
  601. // Year range filtering
  602. search_webofscience({
  603. query: 'machine learning',
  604. year: '2020-2024', // Supports range format
  605. sortBy: 'citations',
  606. sortOrder: 'desc'
  607. })
  608. // Advanced query with filters
  609. search_webofscience({
  610. query: 'blockchain',
  611. author: 'zhang',
  612. journal: 'Nature',
  613. year: '2023',
  614. sortBy: 'date',
  615. sortOrder: 'desc'
  616. })
  617. // Traditional WOS query syntax with field tags
  618. search_webofscience({
  619. query: 'TS="machine learning" AND PY=2023 AND DT="Article"',
  620. maxResults: 20
  621. })
  622. ```
  623. **Supported Search Options:**
  624. - `query`: Search terms (supports multi-topic)
  625. - `year`: Single year "2023" or range "2020-2023"
  626. - `author`: Author name filtering
  627. - `journal`: Journal/source filtering
  628. - `sortBy`: Supported sort field (`date`, `citations`, `relevance`)
  629. - `sortOrder`: Sort direction (`asc`, `desc`)
  630. - `maxResults`: Maximum results (1-100; Starter fetches 50 per page)
  631. - `apiProduct`: `starter` (default) or explicit `expanded`
  632. - `recordView`: Expanded `short` (default) or explicit `full`
  633. - `discoverAccess`: Optional publisher public-page discovery
  634. - `discoverAccessMaxItems`: Discovery bound (1-100; explicit value, deployment default, then 5)
  635. **Supported WOS Field Tags (18 total):**
  636. | Tag | Description | Tag | Description |
  637. |-----|-------------|-----|-------------|
  638. | `TS` | Topic (title, abstract, keywords) | `TI` | Title |
  639. | `AU` | Author | `AI` | Author Identifier |
  640. | `SO` | Source/Journal | `IS` | ISSN/ISBN |
  641. | `PY` | Publication Year | `FPY` | Final Publication Year |
  642. | `DO` | DOI | `DOP` | Date of Publication |
  643. | `VL` | Volume | `PG` | Page |
  644. | `CS` | Issue | `DT` | Document Type |
  645. | `PMID` | PubMed ID | `UT` | Accession Number |
  646. | `OG` | Organization | `SUR` | Source URL |
  647. **Example with Field Tags:**
  648. ```typescript
  649. // Search by PMID
  650. search_webofscience({ query: 'PMID=12345678' })
  651. // Search by DOI
  652. search_webofscience({ query: 'DO="10.1038/nature12373"' })
  653. // Filter by document type
  654. search_webofscience({ query: 'TS="CRISPR" AND DT="Review"' })
  655. // Search specific volume/issue
  656. search_webofscience({ query: 'SO="Nature" AND VL=580 AND CS=7805' })
  657. ```
  658. **🔧 Debugging WOS Issues:**
  659. ```bash
  660. # Enable debug logging
  661. export NODE_ENV=development
  662. # In CI, logDebug is enabled automatically when CI=true
  663. ```
  664. ### Google Scholar Features
  665. - **HTML search adapter**: Uses the Scholar web page, not an official public API
  666. - **Metadata and citations**: Parses titles, authors, abstracts, publication years, and “Cited by” counts
  667. - **Bounded retrieval**: Requests at most 20 results across at most 10 pages with deduplication, dispatch-time pacing, adaptive delays, and bounded retry handling
  668. - **Provider-neutral transport**: Direct Scholar traffic keeps its isolated same-origin HTTPS session; source waits do not occupy the global transport slot, while ScrapingAnt is a bounded fallback for eligible transport/server failures and empty/parse-failed pages
  669. - **Session privacy**: Scholar cookies are created in memory for the exact Scholar origin and are never forwarded to ScrapingAnt or returned in diagnostics
  670. - **No full-text authorization**: PDF/library links remain publisher or institutional links
  671. > **Google Scholar access**: Google’s official help says automated software should respect `robots.txt` and that bulk access is not provided. Direct requests use the standard local proxy aliases (`HTTPS_PROXY`/`HTTP_PROXY`/`ALL_PROXY`, including lowercase forms) when configured. An explicit `SCHOLAR_PROXY` overrides those aliases and is parsed as a complete HTTP(S)/SOCKS proxy URL. With the default transport, the configured ScrapingAnt fetcher is used only as a bounded backup for eligible transport failures, upstream 5xx responses, or no usable/parseable results—not to bypass permission, 403/429, or CAPTCHA responses:
  672. > ```bash
  673. > # Optional explicit HTTP/HTTPS proxy override
  674. > SCHOLAR_PROXY=http://user:pass@host:port
  675. > # Optional explicit TLS-to-proxy endpoint
  676. > SCHOLAR_PROXY=https://user:pass@host:port
  677. > # Optional SOCKS proxy
  678. > SCHOLAR_PROXY=socks://host:port
  679. > ```
  680. > Required packages are loaded lazily (`http-proxy-agent`, `https-proxy-agent`, `socks-proxy-agent`) — install the one matching your proxy type. ScrapingAnt Proxy mode is not used as a transparent substitute: `SCRAPINGANT_PROXY_TYPE` is only the provider proxy ceiling, and the legacy `SCHOLAR_PROXY` path remains independent and outside benchmark acceptance.
  681. ### Semantic Scholar Features
  682. - **AI-Powered Search**: Semantic understanding of queries
  683. - **Citation Networks**: Paper relationships and influence metrics
  684. - **Open Access PDFs**: Direct links to freely available papers
  685. - **Research Fields**: Filter by specific academic disciplines
  686. ### ScrapingAnt Public Fetch Layer
  687. ScrapingAnt is an optional, paid public-page fallback: a key alone does not enable dispatch. Set `SCRAPINGANT_ENABLED=true` to opt in; browser and residential escalation are independent authorizations. The current Scholar `browser:datacenter`-first order is provisional and requires real provider capability validation; otherwise it retains static-first behavior. Without residential authorization, Publisher/Scholar defaults are 50 credits per operation and 10 credits per request; with `SCRAPINGANT_ALLOW_RESIDENTIAL=true`, their defaults become 500/125. Explicit valid limits always win, but residential dispatch still requires `SCRAPINGANT_PROXY_TYPE=residential`; a datacenter ceiling never sends residential traffic. Google Scholar uses the generic `/v2/general` HTML endpoint (not a dedicated Scholar API), while WoS DOI access discovery and Sci-Hub fallback use `/v2/extended`. The layer is never a Clarivate/WoS API proxy. DOI discovery rejects known login/SSO/Clarivate targets before dispatch where the local redirect chain is visible. Local DNS checks cannot prove the remote proxy's own redirect destination, so a discovered link is not an OA, authorization, or copyright determination. Actual usage is taken from response credit headers; local budgets are not provider billing balances. A known permission/security/resource/cancellation/deadline failure does not trigger paid fallback, and completed strategy scopes retain only finite cache metadata rather than raw provider documents. Persistent Scholar blocking remains an external provider limitation.
  688. ### Public-paper and Markdown tools
  689. - `download_public_paper` is a separate, strict MCP tool for `publisher`, `googlescholar`, and `scihub`. Publisher/Sci-Hub use a DOI in `paperId`; Scholar uses a short-lived reference published by the same MCP connection's search result.
  690. - `get_paper_markdown` is explicit opt-in only. It makes at most one static/datacenter `/v2/markdown` request and returns bounded, untrusted Markdown; Markdown is never used for Paper fields, PDF candidate extraction, or implicit search/download calls.
  691. - Scholar references are in-memory only, capped at 256 entries per handler, TTL 300 seconds, non-sliding, and invalidated on handler disposal/restart. Missing, expired, or ambiguous references do not guess a URL or make a provider request.
  692. - Set `SCRAPINGANT_ALLOW_AUTHORIZED_CORPUS=true` plus canonical tokens in `SCRAPINGANT_AUTHORIZED_CORPUS_PLATFORMS` only when a deployment is authorized to attempt restricted pages. This does not forward cookies, Authorization headers, institutional credentials, or MCP-supplied URLs.
  693. - Provider HTML/Markdown is not a PDF authority. The new downloader re-checks public targets, MIME, `%PDF-`, size, cancellation, paths, symlinks, and atomic no-clobber publication. Existing `download_paper` keeps its name, eight-platform schema, and behavior; no Proxy mode or AI Extractor is enabled.
  694. - Offline tests and the default configuration do not send live or paid requests. Live canaries require separate approval, budget, samples, and stop conditions.
  695. ### Sci-Hub Features
  696. - **Opt-in only**: Disabled by default; accepts DOI, `doi:` and `doi.org` forms only
  697. - **Controlled fallback**: Direct mirror lookup first, then bounded ScrapingAnt Extended HTML fallback when enabled
  698. - **Mirror discovery and health monitoring**: Fetches mirror lists from `https://sci-hub.mobi/en/mirrors` and `https://www.ooopn.com/tool/scihub/`, merges optional `SCIHUB_MIRRORS` supplements, then performs max-three-concurrent direct checks with caching and single-flight
  699. - **Explicit states**: Distinguishes not-found, blocked, markup-changed, unhealthy, and transport failures
  700. - **Safe PDF handling**: Public-target redirect checks, MIME/magic/size validation, temporary files, and atomic replacement
  701. - **Compliance notice**: Does not grant access rights or claim copyright/authorization status
  702. ## 📝 License
  703. MIT License - see [LICENSE](LICENSE) file for details.
  704. ## 🤝 Contributing
  705. Contributions welcome!
  706. 1. Fork the project
  707. 2. Create feature branch (`git checkout -b feature/amazing-feature`)
  708. 3. Commit changes (`git commit -m 'Add amazing feature'`)
  709. 4. Push to branch (`git push origin feature/amazing-feature`)
  710. 5. Open Pull Request
  711. ## 🐛 Issue Reporting
  712. If you encounter issues, please report them at [GitHub Issues](https://github.com/your-username/paper-search-mcp-nodejs/issues).
  713. ## 🙏 Acknowledgments
  714. - Original [paper-search-mcp](https://github.com/openags/paper-search-mcp) for the foundation
  715. - MCP community for the protocol standards
  716. ---
  717. ⭐ If this project helps you, please give it a star!

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.