This MCP server provides an interface to the PapersWithCode API, allowing AI assistants to search for academic papers, retrieve their content, and access related code repositories. It simplifies the process of finding research materials and their implementations through a set of specialized tools.
To install the MCP PapersWithCode server for Claude Desktop via Smithery, run:
npx -y @smithery/cli install @hbg/mcp-paperswithcode --client claude
These tools help you find and access research papers:
Search for papers with optional filtering:
result = await search_papers(
abstract="neural networks",
title="attention mechanism",
arxiv_id=None,
page=1,
items_per_page=20
)
Get detailed information about a specific paper by ID:
result = await get_paper(
paper_id="paper123"
)
Extract and read the content of a paper from its URL:
result = await read_paper_from_url(
paper_url="https://arxiv.org/pdf/1706.03762.pdf"
)
These tools allow you to access different components related to papers:
Lists the results for a given paper ID:
result = await list_paper_results(
paper_id="paper123",
page=1,
items_per_page=20
)
Retrieves the tasks associated with a specific paper:
result = await list_paper_tasks(
paper_id="paper123",
page=1,
items_per_page=20
)
Lists the methods related to a given paper:
result = await list_paper_methods(
paper_id="paper123",
page=1,
items_per_page=20
)
Retrieves the repositories linked to a specific paper:
result = await list_paper_repositories(
paper_id="paper123",
page=1,
items_per_page=20
)
List datasets used or referenced in a specific paper:
result = await list_paper_datasets(
paper_id="paper123",
page=1,
items_per_page=20
)
Tools for exploring research domains:
Search for research areas in PapersWithCode:
result = await search_research_areas(
name="computer vision",
page=1,
items_per_page=20
)
Get detailed information about a specific research area by ID:
result = await get_research_area(
research_area_id="area123"
)
List the tasks for a given research area:
result = await list_research_area_tasks(
research_area_id="area123",
page=1,
items_per_page=20
)
Tools for working with paper authors:
Get authors matching a full name:
result = await get_paper_author(
full_name="Author Name"
)
List papers written by a specific author:
result = await list_papers_by_author_id(
author_id="author123",
page=1,
items_per_page=20
)
Tools for accessing conference information:
List conferences with optional filtering by name:
result = await list_conferences(
name="NeurIPS",
page=1,
items_per_page=20
)
Get detailed information about a specific conference:
result = await get_conference(
conference_id="conf123"
)
List proceedings for a given conference:
result = await list_conference_proceedings(
conference_id="conf123",
page=1,
items_per_page=20
)
Get detailed information about a specific conference proceeding:
result = await get_conference_proceeding(
proceeding_id="proc123"
)
List papers presented at a specific conference proceeding:
result = await list_conference_papers(
proceeding_id="proc123",
page=1,
items_per_page=20
)
There are two ways to add an MCP server to Cursor. The most common way is to add the server globally in the ~/.cursor/mcp.json file so that it is available in all of your projects.
If you only need the server in a single project, you can add it to the project instead by creating or adding it to the .cursor/mcp.json file.
To add a global MCP server go to Cursor Settings > MCP and click "Add new global MCP server".
When you click that button the ~/.cursor/mcp.json file will be opened and you can add your server like this:
{
"mcpServers": {
"cursor-rules-mcp": {
"command": "npx",
"args": [
"-y",
"cursor-rules-mcp"
]
}
}
}
To add an MCP server to a project you can create a new .cursor/mcp.json file or add it to the existing one. This will look exactly the same as the global MCP server example above.
Once the server is installed, you might need to head back to Settings > MCP and click the refresh button.
The Cursor agent will then be able to see the available tools the added MCP server has available and will call them when it needs to.
You can also explictly ask the agent to use the tool by mentioning the tool name and describing what the function does.