"Writing about MCP" matching MCP servers:

An MCP (Model Context Protocol) server that gives AI agents live, structured ad intelligence across Facebook, Google, and Instagram — data that no base model can produce from training alone. Powered by Apify actors. Works with any MCP-compatible client: Cursor, Claude, etc.

Enables asking natural-language questions about three published data builds (housing, flights, and a fictional bank), with the agent finding tables, writing SQL, and re-verifying every number against the warehouse before returning an answer.

Enables Claude Desktop to connect directly to Elasticsearch clusters for intelligent log analysis through natural language queries. Users can ask questions about their logs in plain English and get actionable insights without writing complex Elasticsearch queries.

Enables users to ask natural-language questions about a local folder's contents, including file names, sizes, and timestamps, with rollups by type and age and cleanup suggestions for stale or large files. It returns metadata only and cannot read, move, or delete files.

Enables AI assistants to answer questions about French administrative procedures for individuals, professionals and the self-employed using only official service-public.gouv.fr guides. Every answer returns verbatim official text with a citable link and last update date, and flags low-confidence results when nothing relevant is found.

An MCP server for Meertrack that enables users to ask their AI agent about competitors' recent activities, digests, and product changes via natural language.

A small MCP server that gives agents rich context about a YouTube video — its transcript, jump-to-the-moment deep links, metadata, and most-replayed moments — so they can answer questions, summarize, pull quotes, or surface highlights.

Enables MCP clients to look up Bible verses, perform semantic search, and access original-language word study through a local Concord instance, providing accurate citations and honesty about uncertainty.

Enables assistants to search and look up Riftbound TCG cards, list sets, and search or fetch public decklists (resolving stored card UUIDs into real card names) through the community card database's API, so questions about cards absent from model training data can be answered with real data.

An MCP-native research assistant — search backend, intent-routing, and writing pipeline that carries a question through search, verification, and writing, delivering whatever research you need: quick answers, verified sources, or a finished document.

Enables querying Schoox LMS data (users, courses, enrollments, etc.) via natural language using AI assistants, without writing API code.

Enables natural language queries about the technologies behind websites and the companies running them, using the StackScan Tech Lookup API. Includes tools for domain technology lookup, company details, technology usage stats, and batch company lookups.

Lets MCP clients answer questions about website traffic using Plausible Analytics Stats API v2 through read-only tools for totals, trends, breakdowns, and realtime visitors.

Reads and searches local email archives (.mbox or .eml) with full MIME parsing, enabling queries about senders, subjects, dates, and bodies while keeping all data on your machine.

Ask about the Saudi economy in plain English or Arabic. No API key or database required.

Lets AI assistants like Cursor and Claude safely explore your database and answer questions about your data in plain English.

Enables natural-language inquiries about personal listening history and playlist creation with enforced hard constraints and automated repair.

Enables LLM agents to query governed GA4 metrics over MCP, returning definitions, assumptions, and failure modes from a MetricFlow semantic layer without writing SQL.

Provides local academic literature search and writing support by querying OpenAlex, arXiv, and Crossref, downloading open-access PDFs, extracting IMRaD sections, and appending BibTeX references.

Search public Japanese articles about mamakatsu on Mama Rich. Returns article titles, descriptions, and source URLs. No API key required.

Read-only MCP server that exposes the anonym.community privacy research corpus via typed tools, letting agents search 1,478 PII pain points, jurisdiction laws, FAQ entries, and 1,600+ papers with citations. Enables precise, citable answers about privacy law and data-protection issues directly from MCP clients.

Provides coding agents with file inventory, ranked search, location-aware reading, code packing, and note writing over the workspace folder through MCP, without involving a model.

A Model Context Protocol server that retrieves detailed geographic information about IP addresses using the ipinfo.io API, helping determine user location and network details.

Enables small-business AI assistants to answer customer questions about hours, offerings, FAQs, staff, and booking from a single business.yaml config, with industry-specific tools and guardrails.

MCP server that answers questions about creator marketing, influencer pricing, and app growth using a curated corpus of operator quotes with links, and explicitly flags when evidence is thin or absent.

MCP server that lets AI assistants browse and buy ELC Conference 2026 tickets. Enables natural conversation about conference details, ticket availability, and purchasing.

Enables semantic search over ByJG's documentation, letting LLMs answer questions about ByJG libraries and cite the pages where answers were found.

Searches and browses local Claude Code and Cowork session histories stored on your machine, enabling questions about past work without uploading data.

Enables natural language queries to Yandex Metrika analytics, allowing users to ask questions about traffic, sources, and user behavior without dashboards or SQL.

Enables Claude Desktop to answer natural-language questions about customers by exposing search, profile, order history, status filter, and spend statistics tools backed by local mock data, with a single-file swap path to Oracle AI Agent Studio later.

Exposes Victoria's public transport (PTV) GTFS timetable data as MCP tools, allowing LLMs to answer natural-language questions about routes, stops, and departure times.

Enables read-only access to the UK Companies House register, letting models answer questions about company status, ownership, filings, charges, and officers via seven purpose-built tools.

An MCP server that indexes a repository locally and provides keyword, semantic, hybrid, and SQL search tools, enabling coding agents to answer questions about the codebase efficiently without reading files one by one.

FactQL is the harness around the model: ask a question about your data and it runs the query, checks the result before you see an answer, and declines instead of guessing when it can't verify. Every answer ships with the query that ran, the rows it returned and the lineage behind it.

An MCP server that provides offline searchable Microsoft 365 roadmap data and live Azure status, enabling AI agents to answer questions about upcoming features, roadmap changes, and Azure incidents.

MCP server that provides Danish TV guide data, enabling AI assistants to answer queries about TV schedules, channel listings, and programme searches.

Enables MCP clients to search for published academic articles and retrieve detailed metadata, abstracts, citation counts, and related information from the Elsevier Scopus API. Supports citation verification and writing-style analysis grounded in peer-reviewed sources.

Enables AI assistants to search and ask questions about documents stored in AnythingLLM workspaces through MCP tools.

Give your AI coding agent the Roblox DevForum and official creator docs — check if a bug is already known, read the accepted answer, and verify APIs before writing Luau.

Enables pre-submission review of ML papers inside Claude Code or Claude Desktop by checking format and anonymization risks, resolving every reference, finding prior work for each claim, verifying quotes against the PDF, and running methodology and writing checks. It reports findings only, never scores, ratings, or accept/reject verdicts, and never rewrites the author's text.