papserMCP_server
Provides tools for querying the CodeChef VIT papers portal, including searching for question papers by subject, year, exam type, branch, or campus; retrieving full metadata for a specific paper; and accessing a feed of recently added papers.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@papserMCP_server Find recent CAT-1 papers for Data Structures from Vellore campus."
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
CodeChef VIT Papers MCP Server
A standalone Model Context Protocol (MCP) server that exposes the CodeChef VIT papers portal's data to MCP-compatible AI clients (Claude Desktop, Cursor, VS Code Copilot, etc.). It connects directly to the portal's MongoDB database in read-only mode, allowing AI assistants to query university question papers, retrieve full metadata, and fetch feeds of recently added papers.
This server operates purely inread-only mode. It does not contain any write or upload capabilities, authentication flows, or external HTTP server interfaces. It communicates strictly via standard input/output (stdio) transport.
Features
Tool:
search_papers: Search for university question papers using filters like subject (matches both subject name or course code, e.g., "BCSE202P" or "Data Structures"), year, exam type, branch, and campus. Results are capped at 20 documents, displaying a total count notice if more matches exist.Tool:
get_paper_metadata: Fetch the complete metadata details for a specific question paper using its database ID.Resource:
recent-papers: Access a read-only JSON feed of the last 10 papers added to the database.
Related MCP server: arXiv MCP Server
Input Validation & Rate Limiting
Input Validation (Zod)
All tool inputs are validated strictly using Zod before any database queries are executed:
search_papers:subject: Optional string.year: Optional 4-digit integer between1000and9999(e.g.2023).exam: Optional enum matching known exam types:CAT-1,CAT-2,FAT,Model CAT-1,Model CAT-2,Model FAT.branch: Optional enum matching known branch types:CSE,ECE.campus: Optional enum matching known campuses:Vellore,Chennai,Andhra Pradesh,Bhopal,Bangalore,Mauritius.
get_paper_metadata:paperId: Required 24-character hexadecimal MongoDB ObjectId string.
If validation fails, the server responds with a clear MCP error before calling MongoDB.
Rate Limiting
To prevent abuse, the server implements an in-memory Token Bucket rate limiter scoped per session (the stdio connection process):
Limit: Max 30 tool calls per minute.
Behavior: Exceeding the rate limit returns a clean MCP error response (no crash) and logs the event to
console.error.
Setup Instructions
Prerequisites
Node.js (v18 or higher recommended)
npm (v9 or higher)
A read-only MongoDB URI for the portal's database (never use production write credentials)
Installation
Clone or navigate to the repository directory:
cd d:/VIT/Club/papserMCP_serverInstall the dependencies:
npm installCreate your environment configuration file: Copy
.env.exampleto.envand fill in your read-only database URI:cp .env.example .envEdit
.env:MONGODB_URI_READONLY="mongodb+srv://<username>:<password>@cluster.mongodb.net/Papers-staging?retryWrites=true&w=majority"
Build and Run
To compile the TypeScript source into production-ready JavaScript code:
# Build the project
npm run build
# Run the server locally (it will wait for stdio inputs)
npm run startFor development mode (automatically re-compile files as they change):
npm run devWiring into Claude Desktop
To expose this MCP server to Claude Desktop, add the server to your Claude Desktop configuration file.
Configuration Path
Windows:
%APPDATA%\Claude\claude_desktop_config.jsonmacOS:
~/Library/Application Support/Claude/claude_desktop_config.json
JSON Configuration Snippet
Add the following snippet under mcpServers. Make sure to replace the env.MONGODB_URI_READONLY with your actual read-only database credentials:
{
"mcpServers": {
"codechefvit-papers-mcp": {
"command": "node",
"args": [
"d:/VIT/Club/papserMCP_server/dist/index.js"
],
"env": {
"MONGODB_URI_READONLY": "mongodb+srv://<readonly-user>:<password>@cluster.mongodb.net/Papers-staging?retryWrites=true&w=majority"
}
}
}
}On Windows, you can use forward slashes/ in the file path as shown above, or escape backslashes like "D:\\VIT\\Club\\papserMCP_server\\dist\\index.js".
Try Asking Claude
Once successfully connected, you can ask Claude Desktop questions like:
"Can you check if there are any CAT-1 papers for Data Structures [BCSE202P] from the Vellore campus?"
"Find me Vellore campus papers from the year 2023."
"Can you show me the full metadata for the paper with ID 60d5ecb867c2e022f4b2f123?"
"List the last 10 question papers added to the portal."
License
This repository is maintained privately for the CodeChef VIT chapter.
Available Tools
2 toolsget_paper_metadataA
Retrieve complete metadata for a specific question paper by its ID.
Inputs: paperId (required).
search_papersC
Search for question papers from CodeChef VIT Papers Portal using filters. Caps results at 20.
Inputs: exam, year, branch, campus, +1 more.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 2 tools
The two tools have clearly distinct purposes: one searches for papers using filters, the other retrieves metadata for a specific paper by ID. There is no overlap or ambiguity between them.
Both tool names follow a consistent verb_noun pattern: 'search_papers' and 'get_paper_metadata'. The naming is clean and predictable.
With only 2 tools, the server is on the thinner side, but they cover the core search-and-retrieve flow. It feels slightly under-scoped for a question paper portal, but not unreasonable.
The server covers searching and fetching metadata, but it lacks an operation to download or access the actual paper content. Depending on the metadata, this may be workable, but it is a notable gap for a paper-focused server.
Maintenance
Related MCP Connectors
Versioned documentation registry and semantic search for AI tools and coding assistants.
Search Codeforces problems and inspect public problem metadata through the official Codeforces API.
Search the official docs of Claude Code, Cursor, OpenAI Codex and MCP, refreshed daily.
GitHub, npm, PyPI, Stack Overflow, arXiv and OpenAlex paper search as agent tools.
Related MCP Servers
- AlicenseCqualityDmaintenanceEnables real-time search and retrieval of academic paper information from multiple sources, providing access to paper metadata, abstracts, and full-text content when available, with structured data responses for integration with AI models that support tool/function calling.3119AGPL 3.0
- AlicenseBqualityNot gradedmaintenanceEnables AI assistants to search and retrieve academic papers from arXiv through MCP tools, supporting search by various criteria, detailed paper information, category browsing, and PDF content extraction.4102 npm2-
- FlicenseAqualityDmaintenanceEnables searching and retrieving CAIE past-paper questions with filters for subjects, years, and specific topics. It provides LLM-friendly responses including concise text previews and structured JSON data for single or multi-topic queries.71-
- AlicenseAqualityBmaintenanceEnables AI assistants to search arXiv papers, retrieve metadata, browse categories, and read paper text.5MIT