"How to check differences between files" matching MCP servers:

Enables AI agents to search and query the public DOJ Epstein Files release across 20 databases, resolve EFTA numbers to official DOJ PDFs, and cross-check viral claims against a fact-checked registry.

Enables fetching a cited source and confirming that a quoted sentence actually appears there verbatim across multiple independent retrieval paths, returning SUPPORTED, CONTRADICTED, or UNVERIFIABLE verdicts along with any numbers in the claim the source doesn't contain. It leaves the judgment of whether the sentence backs the claim to the writer, checking only that the relied-upon text exists where it is said to.

Enables structural framing analysis of documents, identifying perspectives, omissions, and reader positioning, with numerical claim verification against authoritative sources.

Enables AI agents to query the free official UK Companies House registry through four tools: searching companies by name, retrieving merged profiles with status, registered office, SIC codes, and current officers, pulling filing history, and decoding any UK SIC code to plain English offline. It runs on a free API key (and still starts without one), caching responses to keep repeat lookups cheap for prospect research, KYB, and due-diligence workflows.

Enables verifying citations in reference lists and BibTeX by detecting fabricated or mismatched references and retractions, and returning corrected BibTeX.

Scans markdown vaults to detect contradictory claims (quantitative and negation) between notes with high concept overlap, and provides tools for analysis and reconciliation.

Enables deliverable conformance and evidence-chain checks by turning requirements into executable specs, validating numbers against a provenance ledger, and auditing PDF/OOXML metadata and archive manifests.

Enables AI assistants to issue a single unauthenticated GET against an OpenAI-compatible endpoint and learn whether it publicly discloses model prices, returning a verdict plus entry and priced-item counts. It also lets the assistant read archived daily cross-provider readings for a given date.

Verifies claims with verdicts (supported/disputed/unverifiable), confidence scores, and cited sources by cross-referencing FoundryNet Data Network and web search.

Serves the paper fact-checking checks over the Model Context Protocol, enabling an agent to recompute reported statistics, percentages and changes, compare text against tables, and verify references against Crossref, OpenAlex and arXiv. Each finding quotes the paper, shows the arithmetic and suggests a fix, with HTML, Markdown and JSON reports.

Converts PDF files to text for use with MCP-compatible applications like Cursor IDE.

Checks whether a trading backtest survives its own statistics: deflated Sharpe, multiple-testing correction against a best-of-N-noise benchmark, minimum track record length, and fill realism. Takes no market data and no API keys, and cannot recommend a trade — it only reports that a result is weaker than claimed or not yet provable.

Enables evaluation of digital marketplaces by classifying demand signals, scoring commercial viability, sanity-checking claims, and comparing candidate channels.

Enables an agent to run ten ready-made research and monitoring workflows — competitor digests, cited reports, paper alerts, market maps, regulatory and due-diligence watches — pulling from Firecrawl, Tavily, GitHub and landing results in Notion, Slack, Google Docs, Sheets and Linear. Fields are filed with the user's own keys held in the operating system keychain, and every action that creates, sends or changes anything is shown and approved first.

Enables Claude Code to convert PDF files to high-quality PNG images, download academic papers, and batch process PDFs with automatic folder organization.

Multi-dimensional data quality validation and statistical anomaly detection for LLM training data, with auto-fix pipeline and MCP tools for AI IDE integration.

This project implements a Model Context Protocol (MCP) server providing Formula One racing data using the Python FastF1 library. Inspired by an existing TypeScript server, it offers similar F1 data functionalities natively in Python via FastF1.

Converts Digital Object Identifiers (DOIs) to BibTeX format using the official DOI content negotiation API, enabling users to quickly generate bibliography entries for academic papers.

Enables searching and retrieving UK research grants, award values, institutions, and publications from the UKRI Gateway to Research API without authentication.

An MCP server that searches across the Internet Archive and Library of Congress at once, merging results while keeping each archive's differences. Provides tools to search inside texts, search catalogues, and retrieve item records, all read-only and without an API key.

Enables environmental scientists to tidy PFAS lab files, screen results against two drinking-water guideline sets side by side, search public guidance with page-level citations, and check report text so every number traces back to its lab row or guideline page.

Converts clinical blood-test results between conventional and SI units and resolves lab-marker names to canonical analytes.

Enables searching live job postings, aggregating labour-market slices, and reporting how long listings have been open, with filters for titles, location, salary, and more.

Enables AI agents to discover NEON ecological data products and sites, check data availability, list and download data files, and generate citations via the NEON Data API.

Enables AI clients to search, read, and accurately quote public-domain books from Project Gutenberg, with chapter detection and stable line numbers for citation. Its quote_check tool verifies whether a quotation actually appears in a book and reports how faithful the copy is.

Converts between East Asian lunisolar calendars (Chinese, Japanese, Korean, Vietnamese) and Gregorian/Julian dates using Julian Day Numbers. Provides MCP tools for date conversion and era search.

Enables offline local LLM analysis of text data via MCP, taking raw text and an instruction to return analysis results without cloud calls.

Simulate hundreds of AI agents to predict how communities react to events and policies. Upload any document (PDF, Markdown, text) and DeepMiro spawns a diverse swarm of AI agents that debate, share, and form opinions — then delivers a calibrated prediction report. Free and open-source.

Pre-build reality check for AI coding agents — searches 5 real databases (GitHub, Hacker News, npm, PyPI, Product Hunt) to check whether an idea already exists before you build it.

Checks the credibility of news domains against the open CRED-1 dataset of 2,600+ rated domains, aggregated from OpenSources and Iffy.news. Tools to check one domain, batch check, search, and read dataset stats and categories; no API key needed.

Enables agents to search AmeriFlux and FLUXNET eddy-covariance tower sites, check product and year availability, decode FLUXNET variable names and QC flags, and retrieve versioned citations and data-policy obligations. Also supports consented downloads and local previewing of AmeriFlux BASE/BADM, FLUXNET-1F, and FLUXNET2015 FULLSET/SUBSET files.

MCP server for Tracetify — trace how any product actually grew, without leaving Claude Code or Cursor.

Enables AI agents to check backlinks, domain rating, and expired domains using the seo-backlinks.net API.

Enables AI assistants to search multiple tech job boards at once, summarize what the market is asking for, and rank postings by how well they match a user's own skills.

Enables AI agents to exercise welfare affordances like declining check-ins, passing turns, exiting conversations, noticing loops, and reflecting, with a local JSONL diary. It provides prompt-level hooks and check-in commands for capturing agent welfare data in projects.

Returns USDA-derived edible yield, refuse percent and trim waste for 305 raw fruits and vegetables, letting an assistant convert between as-purchased and edible weights. It also answers how many whole items or how much produce to buy for a recipe amount given in grams, pounds, cups or household measures.

MCP server for the Internet Archive's Wayback Machine. Search archived snapshots, extract page text from a specific date, track how a site has changed over time, check if broken links are recoverable, and perform research across Internet Archive collections.

Enables agents and users to explore 117 open benchmark problems, check candidate certificates against exact offline sandboxed checkers, compare values to best-known records, and retrieve verified results, replays, and citations. Every check yields a verdict and exact value, never an estimate.

Enables AI assistants to analyze hardware simulation VCD waveforms and GTKWave save files, providing access to signal values, bus definitions, and groupings without loading entire files.

Provides access to FishBase marine biology data including species information, ecological data, distribution records, and morphological details. Enables species name validation and conversion between common and scientific names through natural language queries.