paperswithcode_search_papers
Pack: paperswithcode · Connect: https://pipeworx.io/mcp (see Connect below for a single-pack URL)
No MCP client? Call it directly: GET https://gateway.pipeworx.io/v1/tools/paperswithcode_search_papers for the schema, then POST the same URL with its arguments for the data.
Search machine-learning / AI research papers (via Hugging Face Papers, the successor to Papers with Code). Returns arXiv id, title, authors, community upvotes, and a linked GitHub repo when available. Use for “papers on
Parameters
| Name | Type | Required | Description |
|---|---|---|---|
query | string | yes | Search terms (e.g. “mamba”, “diffusion transformer”, “RLHF”). |
limit | number | no | Max papers to return (1-50, default 10). |
Example call
Arguments
{
"query": "attention transformer"
}curl
curl -X POST https://gateway.pipeworx.io/paperswithcode/mcp \
-H 'Content-Type: application/json' \
-d '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"paperswithcode_search_papers","arguments":{"query":"attention transformer"}}}'TypeScript (@pipeworx/sdk)
import { Pipeworx } from '@pipeworx/sdk';
const pipeworx = new Pipeworx();
const result = await pipeworx.call('paperswithcode_search_papers', {
"query": "attention transformer"
});More examples
{
"query": "vision language model",
"limit": 20
}Response shape
Full JSON Schema
{
"type": "object",
"oneOf": [
{
"type": "object",
"properties": {
"query": {
"type": "string",
"description": "The search query used"
},
"total": {
"type": "number",
"description": "Total number of papers matching the query"
},
"returned": {
"type": "number",
"description": "Number of papers returned in this response"
},
"papers": {
"type": "array",
"description": "Array of formatted paper objects",
"items": {
"type": "object",
"properties": {
"id": {
"type": "string",
"description": "Papers With Code paper ID"
},
"arxiv_id": {
"type": [
"string",
"null"
],
"description": "ArXiv ID of the paper"
},
"title": {
"type": [
"string",
"null"
],
"description": "Paper title"
},
"abstract": {
"type": [
"string",
"null"
],
"description": "Paper abstract"
},
"authors": {
"type": "array",
"description": "List of paper authors",
"items": {
"type": "string"
}
},
"published": {
"type": [
"string",
"null"
],
"description": "Publication date"
},
"conference": {
"type": [
"string",
"null"
],
"description": "Conference name"
},
"proceeding": {
"type": [
"string",
"null"
],
"description": "Proceeding identifier"
},
"url_abs": {
"type": [
"string",
"null"
],
"description": "URL to paper abstract"
},
"url_pdf": {
"type": [
"string",
"null"
],
"description": "URL to paper PDF"
}
},
"required": [
"id",
"arxiv_id",
"title",
"abstract",
"authors",
"published",
"conference",
"proceeding",
"url_abs",
"url_pdf"
]
}
},
"count": {
"type": "number"
},
"results": {
"type": "array",
"items": {
"type": "object",
"properties": {
"arxiv_id": {
"type": "string"
},
"title": {
"type": "string"
},
"authors": {
"type": "array",
"items": {
"type": "string"
}
},
"upvotes": {
"type": "number"
},
"published": {
"type": "string"
},
"github_repo": {
"type": [
"null",
"string"
]
},
"url": {
"type": "string"
}
}
}
},
"found": {
"type": "boolean",
"enum": [
true
],
"description": "Present and true on an answer; a miss is the found:false branch."
}
},
"required": [
"query"
]
},
{
"type": "object",
"description": "A miss or an unavailable upstream: found:false with a reason and a hint (the shape check-found-false-schemas.mjs expects).",
"properties": {
"found": {
"type": "boolean",
"enum": [
false
]
}
},
"required": [
"found"
]
}
]
}Connect
Add this to your MCP client config — every tool in the catalog, including this one — or use one-click install buttons:
{
"mcpServers": {
"pipeworx": {
"url": "https://pipeworx.io/mcp"
}
}
}Connect to just the paperswithcode pack
{
"mcpServers": {
"paperswithcode": {
"url": "https://gateway.pipeworx.io/paperswithcode/mcp"
}
}
}See Getting Started for client-specific install steps.