MCP server directory

Deterministic calculation for AI agents: exact-precision math (avoids float64 overflow), date arithmetic (leap years, business days), unit conversion, statistics, subnet math, haversine distance, base conversion, and Levenshtein distance. 15 tools, 100% local, no network calls, no API keys, no telemetry. `npx -y polymath-megablaster-mcp`.

Read-only access to your own Kilter Board climbing logbook: sessions, sends, projects, grade pyramid, hardest sends, month-by-month progression and wall-angle comparison. Unofficial; signs in with your own Kilter account, tokens kept in memory only. `uvx kilter-mcp`

Central Bank of Russia (ЦБ РФ) data — currency exchange rates (daily and historical), key interest rate, inflation, and aggregated macro statistics. Five typed tools backed by an in-memory TTL cache. No API key required.

> Equip your AI with the tools it needs to get the job done Danube is a Tool Marketplace that connects AI agents with tools to hundreds of external services. Further, Danube allows firms to market and market their services to the AI agent economy. Danube exposes 30 tools through the MCP server: Danube works with any MCP-compatible client:

6 tools · create_workflow, execute_tool

The DataFast MCP server lets MCP-compatible clients such as Codex, Claude Code, and Cursor query your analytics and manage your account through natural language. It provides the same functionality, permissions, and safety checks as the REST API and CLI. It uses Streamable HTTP at https://datafa.st/api/mcp. Open Account Settings → API / MCP, choose which websites and permissions the client can access, then create a token. Copy the dft_ token when it appears—you will use it in the next step, and it is only shown once.

6 tools · Check integration status, datafast_analytics_overview

Source-available, local-first video editor for agents: recordings to a finished film. Cut by transcript, hang b-roll off phrases, score, caption, render and master over OpenTimelineIO, ffmpeg, whisper and MLT, then check the render against the edit it came from. Every tool is also a CLI command, plus a browser workspace with an agent pane.

Velog (velog.io) blog server with 23 tools. Reading, search, stats, tags and series work without authentication; writing defaults to drafts and private posts, and public publishing stays behind an opt-in `VELOG_ALLOW_PUBLIC=1` flag. Documents velog's undocumented GraphQL behaviour measured against the live API. 2 runtime dependencies, Node 22.18+, MIT.

Local ledger for development ports: leases a block of ports to each working copy of a project (a clone or a git worktree), writes `.env` files, and resolves service names to URLs, so every parallel coding agent gets its own ports. No daemon and no network listener, so there is little to run and little to attack.

Full double-entry accounting on local GnuCash books: transactions (single and batch), invoices and bills, budgets, investment lots, scheduled transactions, reconciliation, and reports. Modular tool loading — role presets (bookkeeper, investor, freelancer, business) expose just the surface you need, from a 29-tool core up to 87 in all. Multi-currency and multi-book aware, with a plain-text audit trail of every write. Data never leaves your machine. Try it before installing: the browser sandbox opens three populated sample books (USD, CNY, EUR) with nothing to set up; when it clicks, clone and run with `uv`.

Model Context Protocol (MCP) is an open protocol that standardizes how AI applications access external tools and data sources. By connecting LLMs to local development tools, databases, APIs, and other resources, MCP extends their capabilities beyond their base training. The challenge is that running MCP servers locally creates operational friction. Each server requires separate installation and configuration for every application you use. You run untrusted code directly on your machine, manage updates manually, and troubleshoot dependency conflicts yourself. Configure a GitHub server for Claude, then configure it again for Cursor, and so on. Each time you manage credentials, permissions, and environment setup.

5 tools · add_server_to_profile, create_profile

By implementing MCP, Elastic Email allows AI agents to access and control your email operations while maintaining security and proper authentication. - Add contact to a list - Upload and add contacts - Create a list - Create a segment - Create a campaign - Get campaigns - Update a campaign - Pause a campaign - List campaigns - Send emails: both transactional and bulk - Fetch contact, contacts, contact from list - Fetch lists - Fetch templates - Fetch segments - Is ready (health check, connection and readiness check) - Delete contact from list - Delete contact from account

5 tools · add_contact, Build campaigns

Control eSagu functions directly through ChatGPT, Claude, Mistral & others. Adjust prices, process support tickets, manage items – all in chat. Over 20 specialized tools for Amazon, eBay, Kaufland, HelpDesk and Lost & Found - get\_items – search and filter items - edit\_item\_price\_settings – edit min/max/fixed prices - edit\_item\_margin\_calculation – configure margins & costs - edit\_item\_reference\_asins – manage reference ASINs - edit\_item\_strategies – assign strategies & rules - get\_items – search eBay listings - edit\_item\_price\_settings – adjust pricing logic - edit\_item\_search\_term – define competitor search - edit\_item\_strategies – strategies & custom code

6 tools · edit_item_price_settings, edit_item_strategies

The Extentos MCP server (\@extentos/mcp-server\) is an npm package an AI agent (Claude Code, Cursor, Windsurf, Cline) installs once and then uses to add Meta Ray-Ban smart-glasses capabilities to a native iOS or Android app. It exposes a tight set of deterministic tools across 10 categories — discovery, generation, agent configuration, credentials, analytics, guidance, validation, simulation, production-readiness, and documentation — plus a CLI for account linking, telemetry consent, and update checks. This is the agent's operating manual.

6 tools · Check production readiness, Debug with event traces

A WordPress plugin that turns the site it is installed on into a read-only MCP server, so any MCP-capable AI client (Claude custom connectors, and others) can inspect and reason about that site's plugin estate in plain language. Ask "which of my plugins are vulnerable?", "what's bloating my options table?", or "what did deleted plugins leave behind?" — and get answers grounded in the actual state of the site. Three layers inside one plugin, one REST route, no runtime dependencies: - Transport — POST /wp-json/auditra/v1/mcp/{token} speaking MCP over JSON-RPC 2.0 (plain JSON responses, stateless). Both protocol generations are served, decided per request: revision 2026-07-28 (per-request metadata, server/discover, Mcp- header validation) and the initialize-handshake revisions 2025-11-25 / 2025-06-18 / 2025-03-26. Token-in-path auth behind a swappable interface, Origin validation, rate limiting, failed-auth logging. - Collectors — read WordPress directly: inventory, site context, autoload

5 tools · Assess plugin vulnerabilities, Check plugin lifecycle support

The Ferryhopper MCP Server exposes ferry routes, schedules and booking redirects so an AI assistant can discover connections across Europe and the Mediterranean and send users to Ferryhopper to complete their booking. An MCP app (see the official overview) is any client that can consume remote MCP tools and, in some hosts, render UI resources alongside tool results. - 🌍Search ports and connections across 33 countries and 190+ ferry operators - ⛴️Get real-time ferry itineraries with indicative prices - ⚠️Get live disruption alerts across Greece and Europe, including cancellations, vessel changes, port modifications, and weather warnings - 🔗Receive a direct link to Ferryhopper.com to continue the booking flow from the selected trip

3 tools · get_disruptions, get_ports

Query institutional financial datasets through the Model Context Protocol, designed specifically for AI-agent integrations. Clean, structured JSON responses suitable for tools, research copilots, and automation workflows. Clients automatically discover available tools, parameters, and descriptions via the standardized MCP interface. MCP and REST share the same underlying data, auth, entitlements, and stability. One backend, two front doors. Billing-grade usage metering and account-aware access control for production-ready agent deployments.

4 tools · Discover live tools, Get EOD market data

> Connect AI clients to Cal.com scheduling through the Model Context Protocol using the hosted server at mcp.cal.com or a local instance. The Cal.com MCP server wraps the Cal.com API v2 in the Model Context Protocol, letting you manage bookings, event types, schedules, and more through natural language in any MCP-compatible client. The fastest way to get started is to connect your MCP client directly to mcp.cal.com. The hosted server uses Streamable HTTP transport with OAuth 2.1 authentication — your client handles the authorization flow automatically.

4 tools · create_event_type, create_schedule

Create, edit, and publish full-stack apps — database, authentication, and email included — without leaving your AI chat. It runs on your existing Claude or ChatGPT subscription over the Model Context Protocol (MCP), not Floot credits. Add Floot in Claude One click from the Claude connector directory — no URL to copy. Open Floot in the Claude connector directory and click Connect Mention with Floot in a chat and start building Watch the whole flow end to end: add the Floot connector in Claude, then build and publish a real app without leaving the chat.

3 tools · Debug and fix issues, Preview a project

Self-hosted multi-account email server for ChatGPT and Claude: Gmail, iCloud, Fastmail or any IMAP/SMTP mailbox, as many as you like. Search, read, threaded replies, drafts, allowlisted sending and attachments; passwords are encrypted in the browser and the server stores nothing. Remote (OAuth) or Claude Desktop (.mcpb).

Email marketing automation MCP server for Sequenzy. Manage subscribers, lists, segments, templates, campaigns, sequences, transactional emails, analytics, and AI-generated email content. Install: `npx -y @sequenzy/mcp`.

6 tools · add_subscriber, Build email sequences

> Connect Claude, ChatGPT, and other AI tools to your Granola meeting notes via MCP. Query your notes, search transcripts, and get meeting insights in your favorite AI assistants. Connect your AI tools to your Granola meeting notes with Granola MCP. Once connected you can use this to: search through your meeting notes, find specific topics discussed in past meetings, browse meeting folders, extract action items or decisions from your notes, answer questions based on your meeting history and more.

5 tools · get_account_info, get_meeting_transcript

Connect once through the Great Question MCP and run your entire research workflow from any of these tools: create studies, recruit participants, build screeners, analyze results, and pull insights. Spin up studies, recruit participants, and analyze results. The full research workflow from Claude and Claude Code. Run your whole research operation without leaving your IDE: create studies, build screeners, and pull insights. Launch surveys and interviews, manage participants, and dig into results inside ChatGPT.

5 tools · Build surveys, Create interview studies

If you’re an advanced user who already understands MCP concepts and want to quickly connect GTmetrix and start using practical workflows, refer to our setup guide. Learn what the GTmetrix MCP does and how to utilize it in your AI workflow. GTmetrix now has an MCP server that lets you connect your preferred AI tools directly to GTmetrix, enabling AI-assisted performance testing and analysis. Website performance optimization is not a one-off where you run a test, fix issues manually, and forget about it. More and more teams and organizations are working to build intelligent workflows that can analyze, act, and improve continuously.

5 tools · Analyze performance reports, Automate CI/CD validation

Read-only server exposing the agentic-os governance, SDLC, and QE methodology to any MCP host; never writes to your repo and never executes code. Ships install planning and install verification. `npx -y agentic-os-mcp`

Manage Featureflip feature flags from AI agents: create and toggle flags, edit targeting rules, and find stale flags.

Turns the long-running commands a coding agent starts inside Neovim (dev servers, watchers) into regular [overseer.nvim](https://github.com/stevearc/overseer.nvim) tasks, so you can watch and stop them from your own task list instead of them running as invisible children of the agent's shell. `npx -y overseer-nvim-mcp`

An MCP service for deploying HTML content to EdgeOne Pages and obtaining a publicly accessible URL.

5 tools · Deploy a single HTML preview, deploy_folder_or_zip

Unofficial, experimental Mi Fitness connector with local SQLite storage, JSON/CSV exports, and a local stdio MCP server for sleep, activity, workouts, and other available records. Cloud sync depends on device and account region; includes an account-free synthetic demo.

A Model Context Protocol server for generating visual charts using [AntV](https://github.com/antvis).

6 tools · Create diagrams and structured visuals, generate_district_map

A MCP server that provides access to AllTrails data, allowing you to search for hiking trails and get detailed trail information

Maritime vessel tracking via VesselAPI. Search vessels, get real-time positions, ETAs, port events, emissions, inspections, and NAVTEX safety messages.

17 freight calculation and reference tools — ADR dangerous goods, HS codes, LDM/CBM/chargeable weight calculators, duty estimation, airline codes, UN/LOCODE, and more. Free REST APIs + MCP server.

Real-time Lviv (Ukraine) public transport: stop arrivals, live vehicle positions, route geometry, and nearby stop discovery. No API key required.

Keyless remote MCP server for German public infrastructure open data: weather, air quality, traffic, public transit, parking and roadworks across 84+ German cities (DWD, Umweltbundesamt, Mobilithek, GovData). 38 read-only tools.