Skip to content

POWERED BY MCP CONNECT · BY APIDECK

Claude logo

Explore Claude connectors.

Browse MCP integrations reported by the Claude connector directory. Compare endpoints, setup requirements and publisher evidence.

Source: Claude logoClaude connector directory ↗. This is API Tracker’s catalog view, not a list of connections tested by us. Availability in Claude does not establish compatibility with every MCP client.

Compare browser automation MCP servers →

Showing 289–312 of 7,030 resultsPage 13 of 293

Candidate groups, not verified unique MCPs. How grouping works →

1 connection option

Aha!

Official status unknown

Set product strategy, capture customer feedback, prioritize your roadmap, define features, manage releases, and track progress in Aha! software. Bring that information into Claude for full context. Quickly search records, analyze trends, draft documents, and update plans as part of your AI workflow.

1 source

2 connection options

Ahkam — Saudi Court Rulings

Official status unknown

Ahkam (أحكام) is a search engine for Saudi judicial and quasi-judicial rulings that the issuing bodies have published to the public, including the Ministry of Justice, the Board of Grievances, and the committees for tax, customs, and telecom disputes. The connector lets Claude work with this corpus directly: • search_rulings: hybrid keyword and semantic search in Arabic or English, filterable by authority, court, and Hijri year. It returns titles, links, snippets, and metadata. • get_ruling: one ruling by id — metadata and summary, plus its full text for a signed-in Ahkam account (OAuth), within a daily limit. • analyze_facts: send redacted facts and receive an analysis limited strictly to the rulings retrieved, covering material and immaterial facts, the competent court, and the laws cited in those rulings. • rate_result and report_issue: closed-value feedback (no free text) that helps us improve ranking and fix data errors. Every response carries a disclaimer: the output is a research aid, not legal advice, and citations point only to the returned rulings, each linked to its page on ahkam.arbiter.sa. Facts must be redacted before they are sent. The service also masks ID and phone numbers server-side, and stores no facts or queries, only a 30-day usage log without content. Fact analysis is generated by DeepSeek, and every analysis response names the model and its region. Sign in with your Ahkam account (Google or email) through OAuth. Any account can search, preview and read rulings. Fact analysis and feedback tools are available to paid memberships.

2 sources

1 connection option

Ahrefs

Official status unknown

Research keywords, analyze backlinks, track rankings, and monitor brand visibility across traditional search and AI platforms. Discover keyword opportunities, compare competitor performance, audit link profiles, and measure share of voice in LLM responses. Claude combines multiple Ahrefs analyses in a single conversation — asking clarifying questions, cross-referencing data points, and turning complex SEO metrics into actionable insights.

1 source

1 connection option

AI Dispatcher by FieldCamp

Official status unknown

FieldCamp AI Dispatcher brings your field-service operations into Claude. Connect your FieldCamp workspace and Claude can see your live jobs, technicians, daily schedule, and dispatch performance, then act on them: create jobs, run AI dispatch optimization across your team, and accept or reject the AI's suggested assignments — all from chat. It is built for dispatchers and operations teams who want to plan and adjust their day conversationally: "What's on the board today?", "Create a job for 12 Main St tomorrow morning", "Run dispatch and show me the plan", "Accept the suggestion for the Smith job". The connector also surfaces technician and job details, the skills catalogue, EV charging stations and a route map, and multi-stage project schedules. Read tools only retrieve data and never modify it; write tools (create job, run dispatch, accept/reject) make explicit, reviewable changes. Every tool is annotated with read-only or write/destructive hints.

1 source

1 connection option

AI Engine for WordPress

Official status unknown

Connect your self-hosted WordPress site and work on it with Claude. Read and search your posts and pages, write new drafts with real editor blocks, edit existing content, organize categories and tags, manage the Media Library, moderate comments and leave editor notes for your team. Free and unlimited: no daily cap, no plan to buy, no account to create. Your WordPress password is never shared or stored: you approve access on your own site, and you can revoke it there at any time. Connecting takes a minute: enter your site address, approve access in WordPress, and you are ready. If AI Engine is not installed yet, it is installed and set up for you from WordPress.org during that step. You can also choose read-only access when you approve (AI Engine 3.8.7 or later). Everything happens with your own WordPress permissions. New content is created as a draft unless you ask otherwise, deleting only moves items to the trash, and settings, users, plugins, themes and site code are not reachable through this connector. Use cases: - Summarize recent posts - Draft an article with headings and paragraphs - Tidy categories and tags - Set featured images - Moderate pending comments - Leave editor notes on a draft

1 source

2 connection options

AI for Database

Official status unknown

AI for Database connects Claude to the databases in your AI for Database workspace (Postgres, MySQL, SQL Server, MongoDB, Snowflake and more). Ask a question in plain English and get the answer, the SQL that produced it and the rows; inspect schemas; run your own SQL inside a read-only transaction; list dashboards; and create or run workflows that check your data on a schedule and send email or webhook alerts. Queries are read-only: statements that write are rejected by the database. Tools that can send alerts are marked so Claude asks before running them. Plain-English questions use your workspace's AI credits; run_query and the other tools don't. The connector uses a key scoped to your workspace and cannot change billing, members or connection credentials. Revoke it any time under Settings > API Keys.

2 sources

1 connection option

Ai for Work

Official status unknown

AI for Work provides a single, secure connection. Search organizational knowledge with context-aware answers powered by an Enterprise RAG Engine, execute multi-step workflows end to end via the Universal Orchestrator, and leverage ready-to-use solutions like AI for IT, AI for HR and many more alongside custom agents. All interactions respect enterprise security, access permissions, and compliance — making AI genuinely useful at work.

1 source

2 connection options

AI Mail MCP

Official status unknown

One person is often support, sales, billing, legal and HR at once, with a mailbox for each (support@, sales@, billing@, legal@, hr@) and a different voice for each. AI Mail MCP gives Claude all of them as one unified inbox: it sorts what needs a reply, drafts each answer from the mailbox the mail came to, and labels and files the rest. Rules label, archive, forward to verified addresses or draft a reply; they never send on their own. Try it with a 7-day sandbox: you pick a free @myaimail.net address when you connect, no card. New here? Start at aimailmcp.com/docs, or choose the "Test AI Mail on a free sandbox" prompt in Claude. Under the hood it is real email over MCP: search, threads, waiting for a sign-up code or magic link and pulling it out, triage into buckets (needs a reply, bills and receipts, newsletters, notifications), plain-language labels, send, reply in-thread and forward, and your own domain with the exact DNS records to set. Mail from outside senders is untrusted, and the server treats it that way. Every inbound message is scored for prompt injection, and a message scored as one has its body withheld until the agent explicitly asks for it. Outgoing mail passes a guard that checks for credentials, first-time recipients, look-alike domains and misleading links, and holds a credential going to a new recipient until an account owner approves it. AI Mail's own sign-in codes are never readable by an agent. You stay in control. Sign-in is OAuth 2.1 with an emailed code. A connection can only read mail unless you tick "allow sending" or "allow managing the account" on the consent page. Every tool that changes something accepts dry_run, every call is audited, and sending is capped per day. After the trial, pricing is prepaid and per use, with no subscription: keep the sandbox address for $3 (12 months), or top up from $10 to send to anyone. A person pays a Stripe Checkout link; the agent never handles card details.

2 sources

1 connection option

AI NOVA

Official status unknown

AI NOVA is an AI writing assistant from Seoul Economic Daily. This connector lets users generate multiple online headline options from article text and optionally apply their saved MyStyle profile for personalized headline suggestions. Headline generation uses NOVA credits based on actual usage. MyStyle requests are processed asynchronously and return headline candidates that pass AI NOVA's validation checks. AI-generated results should be reviewed by the user before publication.

1 source

1 connection option

AI Passport

Official status unknown

AI Passport is your own shared memory layer for the assistants you choose. Normal saves arrive as proposals in your private inbox. You approve what becomes memory, then grant each app an exact category pass for a chosen duration. AI Passport records content-free activity receipts and can work across assistants without giving one assistant access to every memory.

1 source

2 connection options

AI Retrievability by Norg

Official status unknown

Set up AI Retrievability by Norg without leaving this conversation. Generating a Visibility Audit is free, no trial or card required. Just share your business name and homepage URL, and we'll show you how AI agents currently see your site today. Want your own AI-readable mirror of your site? Jump on a free trial and get started in about 10 minutes, share a few details (you'll add a card via Stripe to activate it, but nothing's charged during the 30-day trial) and you'll have it up and running. Tune it from there using the MCP tools. Norg publishes the mirror (your homepage plus up to 25 more pages found on your sitemap) and keeps it in sync automatically. Once live, get analytics on which AI agents have crawled it, and generate an audit emailed straight to you. We'll remind you by email before the trial ends, and cancelling before then is easy. Send feedback anytime through the MCP. How it works: Websites are built for marketing. Norg mirror sites serve AI the deep context about your business that LLMs rely on, guiding the context, conversation, and consensus AI forms about you. Under the hood, this is edge routing: verified AI agents (ChatGPT, Claude, Perplexity, and others) are served a clean, structured mirror of your homepage, while human visitors keep seeing your site unchanged. The mirror loads fast, runs little to no JavaScript, and stays dense with verified facts and relevant Q&A: everything a model needs for an informed, accurate read on your business. Host with Norg, or self-host as an edge worker on your own CDN: Cloudflare, CloudFront, Fastly, or Bunny. Both modes, and all four CDNs, are covered in the edge routing white paper: https://app.norg.ai/help/edge-routing See live examples: Norg-hosted: https://directory.selleys.com.au. Cloudflare mirror: https://www.cricketforall.com.au, add ?agent=true to the link to see the AI version.

2 sources
Official status unknown

AI Search Kit helps any business become easy for AI answer engines to find, read and cite. audit_ai_readiness reads one public page plus its site's robots.txt, llms.txt and sitemap. It reports, for 12 AI crawlers and control tokens from OpenAI, Anthropic, Perplexity, Google and Apple, whether robots.txt lets each one read the page under RFC 9309 matching. It also reports noindex and nosnippet directives, JSON-LD validity against the schema.org vocabulary and Google's LocalBusiness rules, whether the name, phone and address in the markup appear on the visible page, llms.txt format, the canonical tag and the sitemap. generate_local_business_schema turns business facts into JSON-LD using the most specific of 130 LocalBusiness subtypes, with hours and phone in Google's documented formats, and validates its own output. generate_llms_txt writes an llms.txt in the llmstxt.org format from your facts or from a live site, honoring its robots.txt. check_entity_consistency compares name, phone and address across up to 8 profiles and lists the exact fixes. validate_structured_data checks pasted JSON-LD. Every finding cites the public spec or vendor document behind it, with the date it was read. All tools are read-only, need no account, and call no AI model.

2 sources

1 connection option

AI TeamsMaestro

Official status unknown

Your meetings contain decisions, commitments, and context that your AI assistant can't access. TeamsMaestro changes that by connecting your entire meeting history to Claude. The fastest growing AI note taker for Microsoft users, TeamsMaestro joins your calls on Microsoft Teams, Google Meet, and Zoom, automatically capturing everything so nothing gets lost. Works great for: - Proposals and docs grounded in what was actually discussed in meetings - Client call preparation by surfacing everything from past sessions, then going deeper with Claude - Performance reviews backed by real context and commitments from past conversations - Missed meeting catch-ups with the ability to ask Claude follow-up questions - Engineering workflows enriched with project background, requirements, and technical decisions from planning meetings - Strategic analysis that combines meeting insights with documents and data Instead of copy-pasting transcripts into Claude and hoping it finds the right detail, TeamsMaestro gives your AI direct access to your meeting history. It pulls exactly what it needs, cites the right conversation, and keeps you focused on the work.

1 source
Official status unknown

Vivideo AI Video Prompt Generator turns a rough idea into a detailed, production-ready prompt for any AI video generator. Describe the AI video you want to create, and Vivideo structures your concept into subject, action, setting, camera, lighting, visual style, mood, composition, motion, aspect ratio and duration. It also creates a negative prompt designed to reduce common artifacts in AI-generated videos. Browse 29 curated styles for cinematic AI video, product videos, social clips, talking-head videos, animation, travel, food, fashion, real estate and image-to-video motion. Apply a style using its stable ID, then customize any individual field. Use Vivideo with a text-to-video AI or image-to-video AI workflow to prepare prompts for an online AI video generator, AI video maker, AI video creator, AI animation generator, AI clip generator, AI short video generator or AI marketing video generator. Copy the finished prompt into your preferred tool to generate an AI video from text or an image. The connector is free to use, requires no account or authentication, and does not store your prompts.

1 source

2 connection options

AI Website Security

Official status unknown

AI Website Security reads the public home page of a website the way a visitor does and returns the security basics, each finding with its evidence and what it means for the site. • check_site_security_basics — HTTPS certificate (validity, expiry date, host coverage, protocol), redirection from http to https, security headers (HSTS, Content-Security-Policy, framing protection, nosniff, Referrer-Policy, cookies), mixed content, and whether public DNS filtering resolvers (Cloudflare, Quad9) block the domain — a measured fact: part of the visitors cannot reach the site. Free. • get_security_offer — returns the AI Website Security offers: Record (malware and file-integrity scan with a signed report, 99 €), Record + Restore (cleaning of the core files, 199 €), Watch and Watch + Restore (daily watch for a year), Watch Fleet for agencies — with what each includes and the page where the site owner buys it, accepting the terms and confirming ownership there. Public pages only are read. No account or API key is needed.

2 sources

1 connection option

ai.market

Official status unknown

ai.market is a non-custodial marketplace for AI training data. Sellers list what they have, buyers find what they need, and allAI helps both sides. With this connector you can ask Claude to search published listings, look at a listing's description, columns, row count, update frequency and price, and check your own account, activity and data requests. Just ask in plain language. Claude works with your own ai.market account and only sees what you approve. It can't buy, sell or pay for you. When you want to buy, Claude gives you the link and you finish on ai.market. We never hold the data, and buyers get it straight from the seller.

1 source

1 connection option

AI2SQL

Official status unknown

Ask Claude questions about your own database in plain English. AI2SQL gives Claude your real table and column names, writes the SQL and runs it as a read-only query, so the answer comes from your data. Try it first on demo_shop, a sample store every account can query: "Use AI2SQL to show the top 5 customers by revenue this year in demo_shop" What it does: - Get started: see the databases in your account, what your plan includes, and example questions to ask. - Connect a database: get a link to connect PostgreSQL, MySQL or SQL Server in AI2SQL, including Supabase, Neon, AWS RDS and Azure. Credentials are entered in AI2SQL, never in Claude. - List your databases and read their tables and columns, so queries use real names. - Run a single read-only SELECT or WITH query and return up to 1,000 rows. Writes and DDL are rejected. - Generate SQL from a plain-English request for PostgreSQL, MySQL, SQL Server, SQLite, Oracle, Snowflake, BigQuery and MariaDB. - Fix, optimize, explain and convert SQL you already have. Database on your own machine or network? Save its schema in AI2SQL and Claude writes SQL that matches it. How it works: you sign in with your AI2SQL account over OAuth, and every tool follows your AI2SQL plan. A free account can generate a limited number of queries each day, browse its databases and query demo_shop. Querying your own databases, saving a local schema, and fixing, optimizing, explaining and converting SQL are part of the paid plans, each of which starts with a 7-day free trial. No tool can change or delete data in your database.

1 source

1 connection option

AiApply

Official status unknown

AIApply is your AI career agent for job search, job matching and automatic job applications. It finds jobs that match your preferences and applies for you. Connect your AIApply account to search job listings, track applications and manage Auto Apply without leaving Claude. What Claude can do: * Search current job listings by role, skills, country and remote work, with links to the original listings. * Show matched jobs and applications waiting for your review. * Read the cover letter AIApply generated for an application. * Track the status of job applications AIApply has sent. * Check your job preferences and provide a link to update them. * Check your remaining application allowance and whether Auto Apply is running. With optional permissions, Claude can also: * Show senders and subject lines from your AIApply mailbox. * Apply to a job you found or approve an application awaiting review. * Pause or resume automatic job applications. The connection is read-only by default. You choose each additional permission separately on the AIApply consent screen. Every action that sends a job application requires your confirmation first. Disconnect at any time in your AIApply settings. Claude cannot change your job preferences, write or edit applications, read message contents or delete anything. Your CV file and answers to employer questions are never shared through this connection.

1 source

1 connection option

AIBookSEO Listing Checker

Official status unknown

AIBookSEO Listing Health Checker helps self-published authors and KDP publishers find out why a book listing isn't converting and what to fix first, and estimates sales from any Best Sellers Rank (BSR). No API key needed. Give Claude an ASIN or an amazon.com product link, and the connector reads the public listing and returns: • An overall health score from 0 to 100 • Scores for four areas: title, subtitle, bullet points/description, and reviews • A one-line verdict naming the fastest lever for improvement • Up to six fixes ranked high, medium, or low impact, such as expanding a short title with buyer search terms, removing claims Amazon suppresses, or strengthening the description Claude can then help you rewrite the weak parts in the same conversation. Example prompts: • "Grade the Amazon listing B0CYZHGJ4Q" • "Audit my book at [amazon.com link] and rewrite the subtitle based on the fixes" • "Compare the listing health of these three books in my niche" No account or API key is needed. The connector only reads public listing data and never changes anything on Amazon. It currently supports amazon.com listings. Free use is limited to 10 audits per session per day. Built by AIBookSEO (aibookseo.com), a KDP keyword research and listing optimization platform.

1 source

1 connection option

AIclicks

Official status unknown

AIclicks MCP brings your AI-search brand-visibility data into Claude. It connects to your aiclicks account and exposes the same analytics the aiclicks dashboard runs on, so you can ask Claude how your brand shows up across AI assistants and get answers backed by live data. ▎ Requires an aiclicks account (aiclicks.io) — the server reads from and writes to your aiclicks workspace and does nothing without it. What you can do: • Visibility & mentions — daily visibility share, mention/citation trends, citability, and a per-model leaderboard (ChatGPT, Perplexity, Gemini, AI Overviews, Claude…). • Citations — URL-level citations your brand earns in AI answers, plus the prompts that triggered them. • Competitors — rankings and share-of-voice vs tracked rivals, trends over time, and discovery of rivals you aren't tracking yet. • Sentiment — positive/negative split, trends, themes, and the sources shaping how AI talks about your brand. • Prompts & fanout — the queries aiclicks tracks and the expanded fanout variants behind them. • ChatGPT ads — advertisers, KPIs, leaderboards, timelines, and ad cards seen on your prompts. • GA4 attribution — real site traffic referred by AI assistants: sessions, top pages, locations, devices, and conversions. • Manage — create a domain (brand), add prompts, add competitors. Works across one or many brands. Read tools run freely; write tools (create domain, add prompt, add competitor) always confirm first and need a Pro+ aiclicks plan.

1 source

1 connection option

AIDAN

Official status unknown

AIDAN is the memory layer for everything you save. Instead of saving on each platform separately, you save to AIDAN: any reel, post, video or article worth keeping from Instagram, LinkedIn, X, YouTube and the web. Later, when you are solving a problem, something you saved in the past may already hold the answer. This connector brings your AIDAN library into Claude: just ask "What have I saved on this?" and it brings back what fits. Claude gets read-only access to your library through four tools: find the saves that help with a topic or task you are working on, search your individual saved insights by meaning, browse your most recent saves in order, and open any single save in full. Every result links back to its original source. Read-only, and only ever your own library. Claude can read what you saved to AIDAN and nothing else: it can never add, change or delete a save, and it never sees your browsing history or social feeds. Disconnect anytime from Claude's settings.

1 source

2 connection options

Aide

Official status unknown

Aide is an AI customer service platform. It answers customers on Aide Chat on your website, your support email or Gmail, or inside Zendesk, Front and Gorgias, with the full customer picture: Shopify orders with live tracking and past conversations. When it can't stand behind an answer, or the customer or topic needs a person, it hands off to the right team with a summary, and its draft if you want one. Governance stays with you: shadow mode before go-live, an AI level per topic, approval limits on refunds, a daily AI usage limit and an audit log. New to Aide? Set it up from the chat. Claude checks what's left and does each step with you: your business details, teaching Aide your website and policies, testing the agent, connecting a channel and going live. This connector lets Claude work in your Aide workspace as you, within your role and the approval limits your workspace sets. From a chat you can: - Search your knowledge (help center articles, website pages, products and macros) and read any article. - Find a customer's Shopify orders by email or order number, with status, items, fulfillment and tracking, and check what can be returned. - Find conversations by keyword, status, assignee, topic or helpdesk ticket id, and read their messages, Aide's reply draft and the customer's orders. - Add internal notes, change a conversation's status, priority, assignee or team, and reply to the customer on the conversation's channel. - Refund or cancel an order, and approve or reject order changes waiting for approval. A change above your approval limit waits for an admin. - Read support analytics: volume, AI resolution, handoffs, response times and satisfaction (owners and admins). You sign in with your Aide account and approve what Claude may do: read only, or read and act. Every action is recorded as you. Tools that message a customer, change an order or put the agent live are marked destructive, so Claude asks first. Disconnect anytime under Settings > Developers.

2 sources

1 connection option

AiEngage CRM

Official status unknown

AiEngage CRM connects Claude to your own AiEngage CRM workspace so you can ask questions about your sales and marketing data in plain language instead of building a report. Ask how many contacts you have, what is in the pipeline and at which stage, which campaign performed best last quarter, who on the team logged the most meetings, how a workflow or web form is converting, or what is on your calendar today. Claude reads the answer directly from your live CRM and explains it. The connection is strictly read-only. There is no tool that can create, edit, delete or send anything, so nothing in your CRM can be changed from Claude. You sign in with your own AiEngage login and choose exactly which areas Claude may read — records, reports, analytics, campaigns, segments, lists, workflows, advertising, web forms and more — and which modules within them. Every read runs as you, so Claude sees only the records your CRM permissions already allow you to see, and nothing from any other organization. You can change what is shared, or disconnect entirely, at any time from Settings → API & Connections in AiEngage.

1 source

About the wider MCP catalog

Source listings
166,305
Candidate groups
166,973
Setup-eligible endpoints
30,646

Sources observed 2026-10-24–2026-10-26. Listings can overlap across sources. Groups are identity candidates; setup eligibility is not a successful connection test.

Sources: MCP data from Glama connectors · MCP data from Glama source code · Official MCP Registry · Claude connector directory · API Tracker · publisher documentation · slack-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

    Public Claude connector directory MCP components; desktop extensions excluded. Source tiers are not publisher attestations. Transport and external-client access are not established.

  • API Tracker · publisher documentation

    Publisher-documented remote presets and local servers reviewed by API Tracker. Discoverability does not establish connection readiness.

  • slack-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