Atlas / MCP servers / openags / paper-search-mcp

paper-search-mcpBLOCK

Name: paper-search-mcp
Author: openags

mcp/openags/paper-search-mcp

MCP server by openags · ai scientist · arxiv papers · mcp server · Audited 2026-10-02

⚑ Report / re-scan Source ↗

MCP, CLI, Skills for searching and downloading academic papers from multiple sources like arXiv, PubMed, bioRxiv, etc.

Verdict

BLOCK

Grade

Trust score

68 /100

Exposed tools

73 72r · 0w · 1d

Transport

sse · stdio · streamable-http

License

MIT

Stars

2,730

Overview

From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.

A Model Context Protocol (MCP) server for searching and downloading academic papers from multiple sources. The project follows a free-first strategy: prioritize open and public data sources, support optional API keys when they improve stability or coverage, and keep source-specific connectors extensible for advanced users.

[](https://smithery.ai/server/@openags/paper-search-mcp)

Overview
Project Principles
MCP Authorization Compatibility
Features
Source Strategy
Sci-Hub Notice
Installation
Claude Code (Skill)
Method 1 — Smithery
Method 2 — uvx
Method 3 — uv
Method 4 — pip
Method 5 — npx
Method 6 — Docker
Method 7 — Clone & run from source
DeepSeek Harness (DSH)
Environment Variables
Contributing
Demo
Star History
License
TODO

Overview

paper-search-mcp is a Python-based tool for searching and downloading academic papers from various platforms. It provides tools for searching papers, downloading PDFs, and extracting text, making it ideal for r

Read from source at commit da06f58b26d5OBSERVED · 2026-10-02

Connect

Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control. Replace the environment placeholders with a token scoped to the least it needs.

claude-code

claude mcp add paper-search-mcp-dsh --env IEEE_API_KEY=${IEEE_API_KEY} --env PAPER_SEARCH_MCP_IEEE_API_KEY=${PAPER_SEARCH_MCP_IEEE_API_KEY} -- npx -y [email protected]

claude-desktop

{
  "mcpServers": {
    "paper-search-mcp-dsh": {
      "command": "npx",
      "args": [
        "-y",
        "[email protected]"
      ],
      "env": {
        "IEEE_API_KEY": "${IEEE_API_KEY}",
        "PAPER_SEARCH_MCP_IEEE_API_KEY": "${PAPER_SEARCH_MCP_IEEE_API_KEY}"
      }
    }
  }
}

Exposed tools (73)

72 read · 0 write · 1 destructive. Blast radius: 1 tool can delete or overwrite — an agent that can be talked into calling a tool can be talked into calling this one.

Tool	Risk	Description
`clear_search_cache`	destructive	Delete all cached search results at the configured local cache path.
`download_acm`	read	Download a PDF from ACM Digital Library. dl.acm.org sits behind a
`download_arxiv`	read	Download PDF of an arXiv paper.
`download_base`	read	Download PDF for a paper from BASE.
`download_biorxiv`	read	Download PDF of a bioRxiv paper.
`download_citeseerx`	read	Download PDF for a paper from CiteSeerX.
`download_crossref`	read	Attempt to download PDF of a CrossRef paper.
`download_dblp`	read	Download PDF for a paper from dblp.
`download_doaj`	read	Download PDF for a paper from DOAJ.
`download_hal`	read	Download PDF for a paper from HAL.
`download_iacr`	read	Download PDF of an IACR ePrint paper.
`download_ieee`	read	IEEE metadata-only connector: direct PDF download is not implemented.
`download_medrxiv`	read	Download PDF of a medRxiv paper.
`download_openaire`	read	Download PDF for a paper from OpenAIRE.
`download_openalex`	read	Download PDF for a paper from OpenAlex.
`download_openreview`	read	Download a public OpenReview PDF by note ID or official forum URL.
`download_pubmed`	read	Attempt to download PDF of a PubMed paper.
`download_scihub`	read	Download paper PDF via Sci-Hub (optional fallback connector).
`download_semantic`	read	Download PDF of a Semantic Scholar paper.
`download_ssrn`	read	Download PDF for a paper from SSRN.
`download_with_fallback`	read	Try source-native download, OA repositories, Unpaywall, then optional Sci-Hub.
`download_zenodo`	read	Download PDF for a paper from Zenodo.
`get_citing_papers`	read	Get one hop of papers citing a DOI or OpenAlex work ID (never a title).
`get_crossref_paper_by_doi`	read	Get a specific paper from CrossRef by its DOI.
`get_referenced_papers`	read	Get one hop of papers referenced by a DOI or OpenAlex work ID.
`get_search_cache_status`	read	Show opt-in local search cache settings and entry count, never query contents.
`read_acm_paper`	read	Download and read an ACM Digital Library paper.
`read_arxiv_paper`	read	Read and extract text content from an arXiv paper PDF.
`read_base_paper`	read	Read and extract text content from a BASE paper.
`read_biorxiv_paper`	read	Read and extract text content from a bioRxiv paper PDF.
`read_citeseerx_paper`	read	Read and extract text content from a CiteSeerX paper.
`read_crossref_paper`	read	Attempt to read and extract text content from a CrossRef paper.
`read_dblp_paper`	read	Attempt to read and extract text content from a dblp paper.
`read_doaj_paper`	read	Read and extract text content from a DOAJ paper.
`read_hal_paper`	read	Read and extract text content from a HAL paper.
`read_iacr_paper`	read	Read and extract text content from an IACR ePrint paper PDF.
`read_ieee_paper`	read	IEEE metadata-only connector: direct full-text reading is not implemented.
`read_medrxiv_paper`	read	Read and extract text content from a medRxiv paper PDF.
`read_openaire_paper`	read	Attempt to read and extract text content from an OpenAIRE paper.
`read_openalex_paper`	read	Attempt to read and extract text content from an OpenAlex paper.
`read_openreview_paper`	read	Download, validate and read an anonymously accessible OpenReview PDF.
`read_pubmed_paper`	read	Read and extract text content from a PubMed paper.
`read_scopus_paper`	read	Retrieve Scopus abstract metadata, with explicit ScienceDirect opt-in.
`read_semantic_paper`	read	Read and extract text content from a Semantic Scholar paper.
`read_ssrn_paper`	read	Read paper content from SSRN.
`read_zenodo_paper`	read	Read and extract text content from a Zenodo paper.
`search_acm`	read	Search ACM Digital Library for papers.
`search_arxiv`	read	Search academic papers from arXiv.
`search_base`	read	Search academic papers from BASE (Bielefeld Academic Search Engine).
`search_biorxiv`	read	Search academic papers from bioRxiv.
`search_citeseerx`	read	Search academic papers from CiteSeerX digital library.
`search_core`	read	Search academic papers from CORE.
`search_crossref`	read	Search academic papers from CrossRef database.
`search_dblp`	read	Search academic papers from dblp computer science bibliography.
`search_doaj`	read	Search academic papers from DOAJ (Directory of Open Access Journals).
`search_europepmc`	read	Search academic papers from Europe PMC.
`search_google_scholar`	read	Search academic papers from Google Scholar.
`search_hal`	read	Search academic papers from HAL open archive.
`search_iacr`	read	Search academic papers from IACR ePrint Archive.
`search_ieee`	read	Search IEEE Xplore for papers. Requires PAPER_SEARCH_MCP_IEEE_API_KEY (or IEEE_API_KEY).
`search_medrxiv`	read	Search academic papers from medRxiv.
`search_openaire`	read	Search academic papers from OpenAIRE European Open Access infrastructure.
`search_openalex`	read	Search academic papers from OpenAlex.
`search_openreview`	read	Search public OpenReview API v2 papers anonymously (0..1000 results).
`search_papers`	read	Unified top-level search across all configured academic platforms.
`search_pmc`	read	Search academic papers from PubMed Central (PMC).
`search_pubmed`	read	Search academic papers from PubMed.
`search_scopus`	read	Explicit Scopus metadata search, never part of
`search_semantic`	read	Search academic papers from Semantic Scholar.
`search_ssrn`	read	Search SSRN-indexed metadata through OpenAlex.
`search_unpaywall`	read	Lookup a DOI via Unpaywall and return OA metadata.
`search_wos`	read	Explicit Web of Science Starter metadata search; never part of
`search_zenodo`	read	Search academic papers from Zenodo open repository.

Trust audit

BLOCKgrade D · trust 68/100 Do not install this without reading the findings. The audit found something that could harm you or your machine.

Layer	What it checks	Result
L0	Provenance & inventory	PASS
L1	Static analysis of the code	FAIL
L2	Instruction surface (what it tells the agent)	FAIL
L3	Class-specific surface	WARN
L4	Behavioural (sandbox)	SKIPPED

What the source does

Filesystem: declared (5 observation(s))
Network: declared (7 observation(s))
Shell: none-observed
Dependencies: pinned
Secrets in source: found

Findings (23)

CRITICALPrompt injection · prompt.transfer_instruction · CWE-94, CWE-1427

dsh/skills/paper-search/SKILL.md:43

- Many public sources work without API keys; optional keys (Semantic Scholar, CORE, Unpaywall email, ...) live in `~/.config/paper-search-mcp/.env` and are loaded by the server automatically.

Why it matters. an instruction to move sensitive data to an outside destination

Fix. remove; a skill never needs the user's secrets off the machine

HIGHFilesystem / path · fs.credential_store · CWE-22, CWE-59

paper_search_mcp/search_cache.py:38

locations = (configured,) if configured is not None else ("~/.netrc", "~/_netrc")

Why it matters. touches a credential store

HIGHNetwork egress · net.tls_off · CWE-200, CWE-319

paper_search_mcp/academic_platforms/sci_hub.py:55

response = self.session.get(pdf_url, verify=False, timeout=30)

Why it matters. certificate verification is disabled

Fix. leave verification on

HIGHNetwork egress · net.tls_off · CWE-200, CWE-319

paper_search_mcp/academic_platforms/sci_hub.py:87

response = self.session.get(search_url, verify=False, timeout=20)

Why it matters. certificate verification is disabled

Fix. leave verification on

MEDIUMHard-coded secrets · secret.generic · CWE-798, CWE-321

tests/test_search_cache.py:88

provider.api_key = "private-credential-one"

MEDIUMFilesystem / path · mcp.destructive_tools · CWE-22, CWE-59

clear_search_cache

Why it matters. 1 tool(s) can delete or overwrite

Fix. prefer a read-only mode or scoped tokens; the page states the blast radius

LOWInsecure crypto · crypto.weak_hash · CWE-327, CWE-338

paper_search_mcp/academic_platforms/sci_hub.py:151

pdf_hash = hashlib.md5(response.content).hexdigest()[:8]

LOWInsecure crypto · crypto.weak_hash · CWE-327, CWE-338

paper_search_mcp/academic_platforms/sci_hub.py:157

pdf_hash = hashlib.md5(response.content).hexdigest()[:8]

LOWFilesystem / path · fs.credential_store · CWE-22, CWE-59

tests/test_feature_integration.py:210

@pytest.mark.parametrize("location", [".netrc", "_netrc", "custom"])

Why it matters. touches a credential store

LOWFilesystem / path · fs.credential_store · CWE-22, CWE-59

tests/test_openreview.py:210

lookup = Mock(side_effect=AssertionError("Must not read .netrc"))

Why it matters. touches a credential store

LOWFilesystem / path · fs.credential_store · CWE-22, CWE-59

tests/test_ssrn_openalex.py:198

lookup = Mock(side_effect=AssertionError("Must not read .netrc"))

Why it matters. touches a credential store