MCP Refchecker
A lightweight MCP server that wraps academic-refchecker, letting Claude verify academic citations against Semantic Scholar, OpenAlex, and CrossRef in real time
mcp-refchecker
An MCP server that lets Claude verify academic citations in real time against Semantic Scholar, OpenAlex, and Crossref — catching hallucinated or incorrect references before they end up in your work.
Built on top of academic-refchecker (MIT).
Tool
verify_citation — verifies that a cited paper exists and that its metadata (title, authors, year, venue) matches what was cited.
| Parameter | Type | Required | Description |
|---|---|---|---|
title | string | yes | Title of the cited paper |
authors | string[] | no | List of author names |
year | integer | no | Publication year |
doi | string | no | DOI (e.g. 10.1145/12345) |
arxiv_id | string | no | arXiv ID (e.g. 2301.00001) |
url | string | no | Direct URL to the paper |
Returns JSON:
{
"verified": true,
"url": "https://...",
"matched_paper": {
"title": "...",
"authors": [...],
"year": 2023,
"venue": "..."
},
"possible_match": null,
"errors": null,
"warnings": null,
"info": null
}
Result fields
verified—trueif the paper was found and all provided metadata (year, authors, venue) matches.falseif there is a real metadata conflict or the paper could not be found.matched_paper— the authoritative metadata from the verification source.possible_match— a Crossref fallback match when the exact title was not found but a close variant was (see "Fuzzy fallback" below).errors— hard errors that block verification (wrong year, wrong authors, paper not found).warnings— soft warnings that don't block verification (arXiv v1 vs v2 differences, arXiv preprint vs published venue, incomplete input metadata).info— informational suggestions (e.g., "reference could include arXiv URL").
What counts as an error vs a warning
academic-refchecker returns a flat list of issues with some inconsistency (year mismatches get marked as warnings while author mismatches get marked as errors). This wrapper normalises the output:
- Promoted to hard errors: plain
year/author/venuemismatches where the cited metadata actually differs from reality. These blockverified. - Demoted to warnings: "missing field" errors when the paper was found but the user didn't provide that field in the first place. Missing input metadata is not evidence of a hallucinated citation.
- Kept as warnings: arXiv version differences (v1 vs v2), preprint-vs-published venue notes.
Fuzzy fallback and its limitations
When academic-refchecker reports that a paper could not be verified, this wrapper makes a secondary query to Crossref using fuzzy title matching and fuzzywuzzy.ratio. If a candidate with ≥ 85% similarity is found, it's returned as possible_match with a warning.
What the fuzzy fallback catches:
- Stylistic title variations (case differences, punctuation, word order)
- Minor rewording
- Titles where refchecker's strict comparison rejected an otherwise valid match
What the fuzzy fallback does NOT catch:
- Real typos in distinctive title words (e.g., "Atention Is All You Need")
- Heavily mangled titles
This is a fundamental limitation of free academic search APIs. Crossref, OpenAlex, and Semantic Scholar all do keyword/token-based search — as soon as a distinctive word is misspelled, it simply isn't in the search index, and the real paper won't appear in results regardless of how you post-process them. Catching real typos would require semantic embeddings from a paid API (OpenAI, Voyage, etc.) or a full-text fuzzy search engine, neither of which is exposed by free scholarly data sources.
If you suspect a typo but verify_citation returns unverified, the best workaround is to rewrite the title in the most canonical form you can and try again.
Installation
pip install mcp-refchecker
Or from source:
git clone https://github.com/JonasBaath/mcp-refchecker
cd mcp-refchecker
pip install .
Configuration
Add to your claude_desktop_config.json:
{
"mcpServers": {
"refchecker": {
"command": "mcp-refchecker"
}
}
}
Optional environment variables
SEMANTIC_SCHOLAR_API_KEY— apply for one here for higher rate limits on refchecker's primary verification path.CROSSREF_MAILTO— your contact email, used to opt into Crossref's polite pool for more reliable fuzzy fallback access.MCP_REFCHECKER_DEBUG— set to any non-empty value to print debug logging from the fuzzy fallback path to stderr.
Example with all optional settings:
{
"mcpServers": {
"refchecker": {
"command": "mcp-refchecker",
"env": {
"SEMANTIC_SCHOLAR_API_KEY": "your-key-here",
"CROSSREF_MAILTO": "[email protected]"
}
}
}
}
License
MIT — © Jonas Bååth. Built on academic-refchecker (MIT).
Máy chủ liên quan
UNO: Unified Narrative Operator
A text enhancement tool that transforms story content into rich, detailed narratives using advanced literary techniques and heuristic analysis.
NotebookLM MCP Server
Let your CLI agents (Claude, Cursor, Codex...) chat directly with NotebookLM for zero-hallucination answers based on your own notebooks
AgentHire
AI job search & hiring MCP server with 55 tools. Search jobs, apply, interview, negotiate offers across 20 countries. No account needed to start.
Beancount MCP
Execute Beancount queries and submit transactions to a ledger.
WxO Agent MCP
Simple MCP (Model Context Protocol) server that invokes a single Watson Orchestrate agent remotely. The agent is defined once via environment variables or MCP config. Use this when you want a lightweight MCP that only chats with one agent—no tool management, no agent listing, no flows. Just invoke_agent(message) and get_agent().
Browser Use MCP Server
Automate browser actions using natural language commands. Powered by Playwright and supports multiple LLM providers.
Enzyme
Enzyme turns your Obsidian or markdown vault into a semantic graph that AI can explore. It maps your tags, links, and folder patterns into entities, tracks when you last engaged each thread, and generates catalysts—questions tuned to surface what's latent in your notes.
MetaTrader MCP Server
A Python-based MCP server that allows AI LLMs to execute trades on the MetaTrader 5 platform.
crm-cli
A local-first personal CRM for the terminal — manage contacts, deals, interactions, and tasks from Claude with 18 built-in tools, all stored in a local SQLite file.
Dovetail
Connect AI tools to the Dovetail API for user research and customer feedback analysis.