Awesome Paper MCP
Searches ACM Digital Library publications as one of the aggregated academic sources.
Provides search and retrieval of arXiv preprints: keyword search, PDF download, full-text reading with outlines, LaTeX source access, citation graph lookups, and topic alerts.
Searches the DBLP computer-science bibliography for publication records.
Resolves DOI identifiers (e.g. 10.1145/xxx): get_paper auto-detects DOI-shaped ids and dispatches them to the backends able to resolve the work's metadata.
Runs Google Scholar keyword and advanced searches and retrieves author profile information.
Searches the HAL French open academic archive for papers and metadata.
Searches IEEE Xplore for paper metadata; requires an IEEE API key to be configured.
Searches PubMed and retrieves article full text for biomedical literature queries.
Searches Scopus for indexed publication records and metadata.
Looks up Semantic Scholar paper details, citations, references, author information and paper recommendations; an optional API key lifts the service's 429 rate limits.
Searches SSRN for social-science working papers and preprints.
Searches the Zenodo open research repository for papers, datasets and other deposits.
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., "@Awesome Paper MCP search arXiv and Semantic Scholar for papers on RLHF, then summarize the top 3"
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.
Awesome Paper MCP — Unified Academic Paper Search
One MCP server that unifies 9 upstream paper-search MCP servers into a single endpoint. Point any MCP-capable agent at it and get search, full-text reading, citation graphs, peer-review data and a research pipeline from 40+ academic sources — arXiv, Semantic Scholar, Google Scholar, OpenReview, OpenAlex, DBLP, PubMed, bioRxiv, medRxiv, Crossref, CORE, HAL, Europe PMC, SSRN, Zenodo, CiteSeerX, DOAJ, BASE, OpenAIRE, PMC, ACM, IACR, WoS, Scopus, IEEE, plus conference listings for NeurIPS / ICML / ICLR / IJCAI / CVPR / ECCV / AAAI / PMLR.
14 backends · 148 upstream tools · 1 config entry.
Why a gateway instead of installing 9 servers
Each upstream project pins its own dependency versions (arxiv 2.x vs 4.x,
bibtexparser 1.x vs 2.x, mcp 0.9 vs 1.30, fastmcp 3.x vs 4.x). Installing
them into one interpreter breaks them. This gateway therefore runs each upstream
server as a separate child process on a private stdio pipe and speaks MCP to
it, so no two projects ever share an interpreter.
agent ──stdio/http──▶ paper-mcp gateway ──┬─▶ arxiv (19 tools)
├─▶ multisearch (71 tools)
├─▶ resp (20 tools)
├─▶ openreview (11 tools)
├─▶ openreview_classic (5 tools)
├─▶ scholar (9 tools)
├─▶ google_scholar (3 tools)
├─▶ papers_helper (4 tools)
└─▶ 6 × sercantas research pipeline (6 tools)Related MCP server: Research MCP
Install
One command from a fresh clone (needs uv, git, node ≥ 18, npm):
git clone https://github.com/<you>/awesome-paper-mcp.git
cd awesome-paper-mcp
bash scripts/setup.shThe script creates the Python 3.12 venv, installs the pinned dependency set
(requirements-lock.txt), clones the 9 upstream repos at the exact commits this
gateway was tested against, links them into the same venv with --no-deps
(that is what lets their incompatible pins coexist), installs the Node modules
for the sercantas pipeline, installs the gateway itself, and finishes with an
MCP handshake smoke test. It is safe to re-run.
cd awesome-paper-mcp
uv venv --python 3.12 .venv
export VIRTUAL_ENV=$PWD/.venv PATH=$PWD/.venv/bin:$PATH
uv pip install -r requirements-lock.txt # pinned shared dependency set
git clone <upstream-url> sources/<dir> && git -C sources/<dir> checkout <sha> # ×9, see scripts/setup.sh
for d in blazickjp-arxiv-mcp-server openags-paper-search-mcp monk1337-resp_mcp \
OpenCodice-Research-openreview-mcp anyakors-openreview-mcp-server-equivalent \
silung-scholar-search-mcp elpaca-ai_papers_helper; do
uv pip install -e "sources/$d" --no-deps # pins cannot fight each other
done
( cd sources/sercantas-Academic-Research-MCP-Suite && npm install --no-audit --no-fund )
uv pip install -e . --no-deps # gateway -> .venv/bin/paper-mcpJackKuo666-Google-Scholar-MCP-Server is deliberately not installed: it imports
a sibling module by bare name, so the gateway launches it as a script instead.
Connect your agent
Replace /path/to/awesome-paper-mcp with your checkout everywhere below. If
your client supports a cwd field, the python -m unified_mcp form works; if
it does not (e.g. MiniMax Code), point command at the console script
/path/to/awesome-paper-mcp/.venv/bin/paper-mcp with no args — the gateway
resolves its own paths from the script location.
Claude Code / Claude Desktop
{
"mcpServers": {
"paper-mcp": {
"command": "/path/to/awesome-paper-mcp/.venv/bin/python",
"args": ["-m", "unified_mcp"],
"cwd": "/path/to/awesome-paper-mcp"
}
}
}Cursor (~/.cursor/mcp.json)
{
"mcpServers": {
"paper-mcp": {
"command": "/path/to/awesome-paper-mcp/.venv/bin/python",
"args": ["-m", "unified_mcp"],
"cwd": "/path/to/awesome-paper-mcp"
}
}
}MiniMax Code (~/.minimax/mcp.json or project .mcp.json)
MiniMax Code's MCP settings have no cwd field, so use the console script:
{
"mcpServers": {
"paper-mcp": {
"command": "/path/to/awesome-paper-mcp/.venv/bin/paper-mcp",
"args": []
}
}
}HTTP transport (shared / remote)
.venv/bin/python -m unified_mcp --http --host 127.0.0.1 --port 8765{
"mcpServers": {
"paper-mcp": { "type": "http", "url": "http://127.0.0.1:8765/mcp" }
}
}Generic stdio (Gemini CLI, Cline, Windsurf, Zed, OpenCode, …)
{
"mcpServers": {
"paper-mcp": {
"command": "python",
"args": ["-m", "unified_mcp"],
"cwd": "/path/to/awesome-paper-mcp"
}
}
}The 14 backends
Backend | Source repo | Tools | What it is for |
| 19 | arXiv search, PDF download, full-text reading, outlines, LaTeX source, citation graph, topic alerts | |
| 71 | 40+ sources in one server: PubMed, bioRxiv, medRxiv, Semantic Scholar, Crossref, OpenAlex, DBLP, HAL, CORE, DOAJ, Zenodo, SSRN, WoS, Scopus, IEEE, … | |
| 20 | SERP-free multi-source search + conference listings (NeurIPS/IJCAI/CVPR/ECCV/AAAI/PMLR) | |
| 11 | Reviews, meta-reviews, rebuttals, decisions for NeurIPS/ICLR/ACL ARR/COLM/TMLR | |
| anyakors/openreview-mcp-server | 5 | Bulk conference listing + JSON/PDF full-text export |
| 9 | Semantic Scholar details, citations, references, authors, recommendations | |
| 3 | Google Scholar keyword/advanced search + author profiles | |
| 4 | Local FTS5/BM25 paper library, TOC extraction, in-PDF grep | |
| 1 | Refine a question into a testable hypothesis | |
| sercantas | 1 | Clean and prepare datasets |
| sercantas | 1 | Generate Python/R/JS analysis scripts |
| sercantas | 1 | Execute analysis scripts, capture output |
| sercantas | 1 | Compose APA / ASA / generic reports |
| sercantas | 1 | Coordinate the whole pipeline |
Note on
anyakors/openreview-mcp-server— that URL returns 401 togit cloneand 404 to the GitHub tarball API, so it could not be fetched. Its published content is byte-identical to stepp1/openreview-mcp-server (same tools, same README), which is whatopenreview_classicruns. If the original becomes reachable, replacesources/anyakors-openreview-mcp-server-equivalent.
How to use it
1. Discover what is available
list_sources()Returns every backend, its status (ready / needs_config / unavailable),
the env vars it is missing, and its full tool list.
2. Search across many sources at once
search_papers(query="mixture of experts routing", max_results=10)Fans out to arxiv, multisearch, resp and google_scholar in parallel and
merges the results. Pass sources=["scholar","openreview"] to narrow it.
3. Get one paper by id
get_paper(paper_id="2401.04088") # arXiv id
get_paper(paper_id="10.1145/xxx") # DOIAuto-detects the id shape and tries the backends that can resolve it.
4. Read full text
get_paper_text(paper_id="2401.04088", source="arxiv", max_chars=20000)arxiv handles arXiv ids, multisearch handles arXiv/PubMed/bioRxiv/medRxiv/
Crossref/OpenAlex, openreview_classic exports OpenReview PDFs.
5. Call any specific upstream tool
call_source_tool(backend="arxiv", tool="get_paper_latex", arguments={"paper_id":"2401.04088"})Or, after list_sources() has mounted them, call the namespaced form directly:
arxiv__get_paper_latex(arguments={"paper_id": "2401.04088"})
multisearch__search_pubmed(arguments={"query": "CRISPR off-target", "max_results": 10})
openreview__openreview_get_reviews(arguments={"submission_id": "abc123"})
resp__search_neurips(arguments={"keyword": "diffusion", "year": 2024})Namespaced tools take their arguments as a single arguments object and forward
it verbatim, so every upstream parameter is available. Their descriptions list
each parameter, its type and whether it is required.
6. Run the research pipeline
run_research_pipeline(question="does test-time compute scaling help reasoning?", output_style="APA")Credentials
Everything works with no credentials at all. These only unlock more:
Env var | Backend | Effect |
|
| Removes Semantic Scholar 429 rate limits |
|
| Enables Unpaywall open-access PDF fallback |
|
| CORE full-text search |
|
| IEEE Xplore metadata |
|
| Non-anonymous OpenReview queries |
|
| PDF → markdown parsing |
Set them in the env block of your MCP config, e.g.:
{
"mcpServers": {
"paper-mcp": {
"command": "/path/to/awesome-paper-mcp/.venv/bin/python",
"args": ["-m", "unified_mcp"],
"cwd": "/path/to/awesome-paper-mcp",
"env": {
"SEMANTIC_SCHOLAR_API_KEY": "...",
"PAPER_SEARCH_MCP_UNPAYWALL_EMAIL": "you@example.com"
}
}
}
}Layout
awesome-paper-mcp/
├── unified_mcp/ # the gateway (the only code in this repo)
│ ├── server.py # FastMCP gateway + smart tools + namespace mount
│ └── backends/
│ ├── registry.py # the 14 backend definitions
│ └── proxy.py # raw JSON-RPC stdio bridge to each child
├── scripts/setup.sh # one-command bootstrap (clones + installs everything)
├── requirements-lock.txt # pinned shared dependency set (144 packages)
├── mcp-configs.json # ready-to-paste configs for 10 MCP clients
├── pyproject.toml # console script: paper-mcp
├── sources/ # 9 upstream repos — fetched by setup.sh, not committed
├── .venv/ # Python 3.12 environment — created by setup.sh
└── README.mdThe upstream projects are not vendored: sources/ is git-ignored and
populated by scripts/setup.sh at the commits recorded in that script, so they
keep their own history, licenses and update cadence.
Design notes
Why raw JSON-RPC instead of mcp.client.stdio? The installed mcp 1.30.0
client deadlocks inside stdio_client + ClientSession.initialize() on this
platform (an anyio 4.15 task-group / cancel-scope interaction). All upstream
servers answer correctly to raw newline-delimited JSON-RPC over stdio, which was
verified against every backend first, so backends/proxy.py speaks the protocol
directly. If a future mcp release fixes the client, swapping in the official
client is a drop-in change confined to proxy.py.
Why lazy startup? A backend is only spawned when first used, so one slow or
broken source never delays the handshake or blocks the others. --no-lazy
warms everything at boot instead.
Proxy handling. If the host exports an http_proxy pointing at a loopback
port that is not listening, the child processes detect it and fall back to a
direct connection instead of failing every request with ProxyError.
License
MIT for the gateway code in unified_mcp/ and everything else in
this repository. The upstream projects under sources/ are downloaded at
install time and keep their own licenses — nothing here redistributes them.
Available Tools
6 toolscall_source_toolA
Call any tool from any backend by name.
This is the escape hatch for backend-specific features that have no
cross-source wrapper (e.g. openreview reviews, arxiv LaTeX source,
resp conference listings). Use list_sources() to discover tool names.
Args: backend: Backend name from list_sources(). tool: Tool name as listed by list_sources(). arguments: Tool arguments as a JSON object.
Inputs: tool (required), backend (required), arguments. Returns: result
get_paperA
Fetch metadata / abstract for one paper by its id.
Args: paper_id: arXiv id (2401.04088), DOI (10.1234/xyz), OpenReview forum id, or a Semantic Scholar paper id. source: Force a specific backend. Omit to auto-detect.
Inputs: paper_id (required), source. Returns: result
get_paper_textA
Download a paper and return its readable text.
Args:
paper_id: arXiv id, DOI, or backend-specific id.
source: Backend to use. arxiv (default) supports arXiv ids;
multisearch supports arXiv/PubMed/bioRxiv/medRxiv/Crossref/OpenAlex/...;
openreview_classic exports OpenReview PDFs.
max_chars: Truncate the returned text to this many characters.
Inputs: paper_id (required), source, max_chars. Returns: result
list_sourcesA
List every paper source available through this server, with its capabilities, tool names and any missing credentials.
Call this first to decide which backend to use.
Inputs: no named parameters. Returns: result
run_research_pipelineB
Run the full academic research pipeline on a research question.
Stages: refine the question into a hypothesis, generate analysis code, execute it, then compose a structured report. Uses the sercantas Academic-Research-MCP-Suite backends.
Args: question: A vague or broad research question. output_style: APA, ASA or generic.
Inputs: question (required), output_style. Returns: result
search_papersA
Search for papers across one or many sources at once.
Default sources are the ones that need no credentials: arxiv, multisearch (Semantic Scholar / Crossref / OpenAlex / DBLP / PubMed ...), resp and google_scholar.
Args: query: Keywords or a paper title. sources: Backend names from list_sources(). Omit for the default set. max_results: Results per source. year_from: Only papers from this year onwards (where supported).
Inputs: query (required), sources, year_from, max_results. Returns: result
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 6 tools
The core wrappers (list_sources, search_papers, get_paper, get_paper_text) target clearly distinct steps, and the metadata-vs-full-text split is explicitly documented. However, call_source_tool is a generic escape hatch that can perform the same operations as every wrapper, so an agent could plausibly misselect it instead of search_papers or get_paper without reading carefully.
All six tools use snake_case with a leading verb (list_, search_, get_, call_, run_), giving a predictable verb_noun pattern throughout. The paired get_paper / get_paper_text naming even signals their relationship clearly.
Six tools is lean for a multi-backend aggregation server and each has a distinct role: discovery, search, metadata, full text, escape hatch, and pipeline. Nothing feels redundant or padded.
The surface covers the full lifecycle: discover backends, search across them, fetch metadata, retrieve full text, and reach any backend-specific feature (reviews, LaTeX, conference listings) via call_source_tool. Citation-graph traversal is not a dedicated wrapper but is reachable through the escape hatch, so there are no real dead ends.
Maintenance
Related MCP Connectors
arXiv and OpenAlex/Crossref paper search (an open Google Scholar alternative) as agent tools.
Search arXiv/Semantic Scholar/OpenAlex + medical evidence (PubMed/Europe PMC) + LaTeX/PDF tools.
Scholarly search: OpenAlex, Crossref, arXiv, OpenCitations and PubMed in one endpoint.
Research paper search with real citations and reference formatting for AI assistants
Related MCP Servers
- FlicenseNot gradedqualityDmaintenanceEnables AI assistants to search across multiple academic databases (PubMed, arXiv, bioRxiv, medRxiv, Semantic Scholar) through a unified interface. Supports advanced filtering, metadata retrieval, PDF downloads, and comprehensive research workflows with citation analysis.5-
- AlicenseNot gradedqualityDmaintenanceEnables LLMs to search, analyze, and summarize academic research papers in real-time from arXiv, Semantic Scholar, and PubMed. Provides automatic deduplication, citation analysis, and BibTeX generation across multiple research databases.22 npmMIT
- FlicenseNot gradedqualityCmaintenanceEnables searching for academic papers and preprints across multiple platforms including Semantic Scholar, arXiv, PubMed, and CrossRef. It provides access to research records, DOI lookups, and journal metadata through a unified interface deployed on Cloudflare Workers.2-
- AlicenseAqualityBmaintenanceUnifies academic literature retrieval from multiple sources like Scopus, ArXiv, PubMed, and Google Scholar into a single standardized API for LLM agents.2934 PyPI11AGPL 3.0