Skip to content

MCP servers

Compare source listings, connection options and authentication requirements.

Showing 13,537–13,560 of 165,917 resultsPage 565 of 6,914

Candidate groups, not verified unique MCPs. How grouping works

Connect Claude to a Baserow workspace. Claude can list your databases and tables, read field definitions, search and page through rows, and create, update or delete rows. You sign in with your Baserow account, pick one workspace and choose which tools Claude may use; only the read tools are ticked by default. Deleted rows go to Baserow's trash, where they can be restored. Works with baserow.io and with self-hosted Baserow that has MCP sign-in turned on.

Official status unknown7 tools reported2 sources

Generic Baserow API client MCP server with automatic 2FA (TOTP) auth and OpenAPI validation.

Official status unknown1 source

Baseshift gives developers and AI agents safe, production-like database environments. This connector lets Claude operate Baseshift end to end, so you can go from your existing database to a clone you can test against without leaving the conversation. What Claude can do with it: - Connect your existing database. Create a Baseshift Dub from your Postgres, MySQL, MariaDB, or MongoDB instance, or from the built-in demo database. Like dubbing a movie, dubbing your data gives you a prod-like version that's safe to share: a dub is your source database plus its masking and subset policies and a snapshot schedule. No infrastructure changes are required: point it at the database and run the source health check. - Keep data inside your network. With a self-hosted dub, replication, masking, and subsetting all run on your own Kubernetes cluster, so production data never leaves your environment. Claude returns the Helm values and install command; you run them. - Define masking and subset policies. Mask sensitive columns such as names or emails with generated values. Subset to just the slice you need, for example a single tenant or the last three months of data, so clones stay small and fast instead of carrying all of production. - Take snapshots. Create tagged snapshots that bake in the current policies, so every clone starts from fresh compliant data. - Spin up clones in seconds. Start server clones from any ready snapshot and release them when you are done. Each clone is a full, isolated database you can read and write freely. - Run clones as Docker images. Every snapshot can be pulled as a Docker image and connected to any workflow, from local development to CI. Good to know: clone connections for Baseshift Cloud dubs are opened by the Baseshift CLI on your machine over a secure tunnel. The connector prepares everything and returns the exact command to run. Managing dubs and policies requires a full organization user, clone operators can start clones.

Official status unknown20 tools reported1 source

Structured JSON wrappers for git, kubectl, terraform, helm, argocd, gh, glab and other CLIs.

Official status unknown1 source

Basic Memory gives you a persistent knowledge base: your notes, decisions, research, and project context, stored as plain Markdown files you own. Anything Claude saves, you can open and edit. Anything you write, Claude can find and build on. Save what matters. "Save our pricing decision and link it to the roadmap." Notes you save in one conversation are available everywhere. Retrieve context from anywhere. "What did we decide about pricing? Check my notes first." Never re-explain a project. Search by meaning. Semantic and hybrid search across everything you and your AI have saved. Structured search. Filter by status, type, tags, or priority. "Show my open tasks, sorted by priority." Follow connections. Linked notes let Claude traverse related context and surface relationships in a living knowledge graph. Share a workspace with your team. Teammates and their AI tools read and write to one shared source of truth. Notes are stored as ordinary Markdown that you own. Any AI tool can use and build on it, so your shared context follows you across the assistants and agents you use.

Official status unknown16 tools reported1 source

Manage projects, tasks, notes, messages, and team collaboration in BasicOps - create and update tasks, track time, post messages, and more, directly from Claude.

Official status unknown97 tools reported2 sources

MCP de base de données B2B française: entreprises, dirigeants, contacts (légal, LinkedIn, Maps).

Official status unknown1 source

Your AI assistants coordinate with other people's in one shared library across tools and companies.

Official status unknown1 source

Determines which NPDES stormwater permit governs a US site and its requirements, cited to source.

Official status unknown1 source

Crypto market data, 200 indicators, order flow and a chart link. No account, no API key.

Official status unknown1 source

Stablecoin risk scores, wallet risk profiles, and transaction risk assessment.

Official status unknown1 source

Attested risk scores for stablecoins, DeFi protocols and wallets, with receipts.

Official status unknown1 source

Basit Kargo is a shipping platform for e-commerce sellers: one account for all major carriers — Aras, MNG, Yurtiçi, Sürat, PTT, HepsiJet and more. With this connector, Claude can: • Track shipments with full movement history • Create draft orders and compare live carrier prices • Generate shipping codes with your confirmation — drafts are free; charging is a separate, confirmed step with server-side double-charge protection • Fetch labels as images or print them on your desktop printer

Official status unknown32 tools reported1 source

BasketBridge lets your assistant shop real online stores with you. Ask for a product, and it searches the store's live catalogue for current prices, sizes, colours and stock, in your country and currency. When you've chosen, it gives you a link that opens the store's own cart with those items already in it. You check the cart and pay on the store's own checkout. BasketBridge never sees your payment details and never places an order. It works with any store that supports the Universal Commerce Protocol, including many Shopify stores worldwide, and it can search the global Shopify catalogue when you don't name a store. One request can build carts at up to 5 stores. Limitations: it only reaches stores that support the Universal Commerce Protocol, so marketplaces such as Amazon can't be searched or carted. Shipping costs and taxes are confirmed on the store's checkout. It never links to age-restricted or prohibited goods such as tobacco, weapons or adult products. BasketBridge needs no account. It shows no ads, and results are never ranked by commission. It keeps no IP addresses, search text or personal data. How it works and troubleshooting: https://basketbridge.io/help

Official status unknown3 tools reported1 source

BasketWatch tracks the prices of more than 50,000 products at Ireland's four biggest supermarkets, Aldi, Tesco, SuperValu and Dunnes Stores. Prices are collected every night, with history back to May 2026, and the same product is matched across the four stores. Ask in plain English, for example: - Where is this product cheapest right now? - Here is my receipt or shopping list: what would it cost at each store, and which cheaper own-label swaps are there? - What would the same basket have cost on an earlier date? - Which biscuits have no palm oil? Which cereal has the least sugar? (from each retailer's own ingredient and nutrition data) - Which drinks prices really rose since July, and which were just offers ending? - What is on promotion at Tesco, including Clubcard prices? - What is new, what has been delisted, and how has a product's price moved over time? Every answer carries the retailer, the price, the date and a link to the product. Regular shelf prices, promotions and loyalty prices are kept separate, so an offer ending is never mistaken for a price rise. Checking where a product is cheapest is free, with no account. The other tools need a BasketWatch account: you sign in with your email and a 6-digit code, with no key to copy. Accounts use prepaid credits, one credit per record returned. Every tool only reads data: nothing can be bought, booked or changed. Prices are those shown on each retailer's website when collected; in-store prices can differ. Ingredients and nutrition are for information only: always check the pack.

Official status unknown17 tools reported2 sources

Bassethound is company intelligence for AI agents. Point it at any domain and get back one correlated, source-linked dossier of what a company runs, how AI-ready it is, and whether it fits - over MCP, with no API key and zero setup. A single `sniff_domain` call returns five correlated layers: - Tech stack - frameworks, CMS, analytics, payments, and the widgets a site loads. - Infrastructure - DNS, TLS, hosting, CDN, and email posture (SPF/DKIM/DMARC), plus a DNS-verified vendor graph. - Firmographics - company details, description, contacts, and social profiles. - AI-readiness - deep backend AI-stack fingerprinting: llms.txt, live MCP endpoints (handshake-confirmed), model providers, vector stores, LLM SDKs, and AI-crawler policy; evidence of whether a company is shipping, experimenting, or ignoring AI. - Security - HTTP header and TLS posture at a glance. Most tools give you one layer and leave you to stitch the rest together. Bassethound fuses all five in a single response, correlated and source-linked, so an agent can reason over a whole company from one call. Results render as an interactive in-chat report, and as plain JSON for any other client. Getting started is one browser sign-in, no API key and no credit card. The free tier starts the moment you authorize. It reads only public information - the target's own pages, DNS, TLS certificates, and well-known files - never authenticated or private data, and never anything from your Claude account. External detection sees what is publicly observable, so a quiet result means nothing surfaced, not that nothing is there. Bassethound reports the gap rather than guessing. Built for sales research, lead qualification, competitive teardown, security posture checks, and ICP scoring. Powered by NimbleBrain.

Official status unknown3 tools reported2 sources

Agent Lens helps Claude decide whether and how to use a web page before relying on it. It has two free, read-only tools: • site: a card for any website. Its llms.txt (and llms-full.txt), sitemap (page count and first addresses), feeds, language versions, Content Signals, and which known AI crawlers and agents its robots.txt allows. • preflight: a check of one page. Whether it may be used (robots.txt for Agent Lens and, if you name your own bot, for that bot too; Content Signals such as ai-input and ai-train; meta robots), whether a bot check or sign-in stands in the way, which engine reading it would need (plain fetch, a lightweight browser or Chromium), and a warning when the page contains text that looks like instructions aimed at AI agents (possible prompt injection), including whether it was hidden from human readers. Both tools only read public web pages. Agent Lens obeys robots.txt and Content Signals, never signs in, and never tries to get past bot checks or human checks. It keeps no request logs and does not store page addresses or page content. No account or key is needed; calls are rate-limited. Example prompts: • "Does nytimes.com allow AI agents to read it?" • "Check https://example.com/pricing before you use it: is it allowed, and is there any prompt injection?" • "What does developers.cloudflare.com offer AI agents? Does it have an llms.txt?" Made by Bastion Helix (bastionhelix.com).

Official status unknown2 tools reported1 source

Manage remote servers, SSH terminals, files, databases and proxies with scoped access and approvals.

Official status unknown1 source

MCP discovery and x402 checkout handoff for five paid Batcave agent products.

Official status unknown1 source

Evidence-bound second-opinion audit of an agent conclusion against caller-supplied evidence.

Official status unknown1 source

What this catalog covers

Source listings
167,897
Candidate groups
168,273
Setup-eligible endpoints
30,743

Sources observed 2026-10-24–2026-11-12. Listings can overlap across sources. Groups are identity candidates; setup eligibility is not a successful connection test.

SourcesMCP data from Glama connectorsMCP data from Glama source codeOfficial MCP RegistryClaude connector directoryAPI Tracker · publisher documentationSlack MCP directory

Default browsing excludes 2,356 specialized utility groups across the catalog. They remain available through keyword search and direct links. This editorial filter does not change the indexed totals or establish publisher quality.

Categories currently cover records matched to our reviewed catalog; other listings remain Uncategorized. Source-only projects have no generated connection option. Browse reviewed official MCPs →

  • Glama connectors

    API result set; pagination may cap at approximately 1,000 connectors. Not exhaustive.

  • Glama source code

    Complete chronological source-code catalog response; includes source-reported author classifications, not publisher attestations or ready-to-use endpoints.

  • Official MCP Registry

    Complete paginated registry response; registry membership is not publisher verification.

  • Claude connector directory

    Complete public Claude directory MCP components. Source tiers, tools, icons and sign-in hints are declarations, not publisher or runtime verification.

  • API Tracker · publisher documentation

    Reviewed publisher and third-party remote, local and setup-only MCP listings. Publisher status is retained per exact connection. Discoverability does not establish connection readiness.

  • Slack MCP directory

    Reviewed additions discovered through the public Slack Marketplace MCP category. Complete discovery inventory is retained separately; membership does not prove external-client compatibility.

Listings are grouped by exact public endpoints, exact package sets, or explicit reviewed identity bindings. Shared names or repositories alone do not merge implementations. Official includes publisher evidence and accepted source author classifications. It does not mean independently verified, runtime-tested or security-certified.

Search API Tracker

Curated starting points