Explore Claude connectors.

Temp Mail gives Claude its own disposable inboxes, so an agent can complete any flow that sends an email without involving your real mailbox. Create a temporary address on demand, check it for incoming mail, and read the message — including the verification code, magic link, or attachment it carries. Addresses and their messages can be deleted the moment you're done. Typical uses: • Sign up for a service end to end and confirm the account, without a human pasting the code • Exercise your own signup, password-reset, or email-verification flows from an agent or a CI run • Inspect exactly what an email looks like on arrival, including raw RFC-2822 source and headers • Fetch a PDF, CSV, or image attachment that was emailed to you and work with it directly Nine tools: list_domains, create_email, delete_email, list_messages, get_message, get_message_source, delete_message, get_attachment, get_rate_limit. Six only read data; create_email adds an address; delete_email and delete_message remove data and are annotated as destructive. Requires temp-mail.io account. You connect over OAuth 2.0 — Claude never sees your password, and you can revoke access at any time from your account settings. Tool calls draw on the same monthly request quota as the temp-mail.io REST API, and get_rate_limit reports what's left at any point.

Easily build your small business a stunning website or app in ~24 minutes. Unlike most AI-website builders, Vibe Otter was created by a professional small business web design firm who deeply understands what quality and features owners need to feel comfortable putting in front of customers. Vibe Otter pages automatically come with captivating heroes that will draw your prospects attention and drive them further down your page to carefully crafted calls to action. All pages automatically come with a fully configured free form service, normally a $5-10 separate subscription. All pages are automatically optimized for SEO and contain all required SEO files. Vibe Otter can do more than websites though - it can build literally any app you want. Many customers come to us for their initial website but quickly realize they can replace all of their booking, CRM, and more SaaS subscriptions with a few prompts in Vibe Otter. Examples are - many salons replace their Vagaro subscription with their own custom booking on Vibe Otter. Many services just make their own CRM's. The power of Vibe Otter as an App builder is it builds the entire full stack app for you one-shot. This means that you get authentication, a database, storage, and compute all from one prompt. Most app builders require you to architect all of this out and connect everything over the course of a week. More importantly, doing it yourself requires sophisticated knowledge no small business owner has, and often creates security and SEO vulnerabilities you may realize are serious issues months after you create an app. Our many customers agree it makes a lot more sense just to have these complex configurations decided for you, so that you can focus whats important: making sure the customer flow is exactly what you need for your business.

wawesome runs JavaScript Functions on WebAssembly, and this server lets Claude manage them in your workspace. Claude can: - deploy a Function together with the pages and files it serves, in one version - run a Function, or fetch one of its paths, and read the result - list versions and roll back to an earlier one - list invocations, read their logs, and explain the latest failure - set environment variables (values can never be read back) - attach a custom domain and check where its certificate got to - show this period's usage against your plan Every call acts inside the one workspace you approve when you sign in. On the consent screen you pick what Claude may do: read only, ship and read back, or build. You can also limit it to certain Apps. Nothing reaches outside that workspace. New to wawesome? Signing in through the connector creates your workspace. There is nothing to install.

Spreadsheets come back wrong more often than they come back broken. Most tooling reads a workbook as a grid of values, which works until the file has formulas — and then two things go wrong quietly. Values come back as whatever was last cached, which may be stale or missing entirely. And a round trip through a re-render drops everything that wasn't a cell value: charts, pivots, conditional formats, data validations, named ranges, styles. Neither failure announces itself. The file opens, the numbers look plausible, and the analysis runs against them. xlsx-for-ai recalculates in its own engine across 382 Excel functions, including dynamic arrays and XLOOKUP/FILTER, returning freshly computed values rather than cached ones. Edits are a structural splice into the existing workbook, so anything you didn't touch passes through untouched — and every write names the engine that produced the bytes, so you always know which path served you. 56 tools cover the round trip: read, inspect, recalculate, write, diff, clean, redact, and repair broken external references. It also opens formats that can't be uploaded at all — .xls, .xlsb, .ods, .numbers, Lotus, Quattro Pro — which is how an agent gets a legacy file in. Workbooks up to 100MB and CSVs up to 200MB are handled server-side, past the point where loading a file directly stops being practical. Parsing is hardened against XXE injection, zip bombs, and entity expansion. Formula evaluation is sandboxed with network functions disabled, and recalculation runs entirely in our own path — no third-party formula engine sees your data. Free accounts include a monthly file allowance; higher volume requires a paid plan.

APIANT is an integration platform you drive from Claude. Describe the integration you want in plain English, such as "when a deal closes in HubSpot, create the customer in QuickBooks and post to Slack", and Claude builds it as an APIANT automation: it picks the apps, maps the fields, connects your accounts, and tests every branch before turning it on. What you can do with this connector: - Build, edit, test and turn on automations between your apps, with triggers from polling, webhooks, schedules, forms and chat. - Diagnose a failed run: read each step's input, output and error, then retry, replay or restart from a step. - Add an app that is not in the catalog: Claude reads the vendor's API docs and builds a private connector, action or trigger, tested before use. - Apply the comments you leave on an automation's diagram in the APIANT web app. Every automation is a logic diagram you can read and change in the APIANT web app, not a black box of code. Claude loads APIANT's step-by-step skills over the same connection, so there is nothing to install. Sign in with your APIANT.ai account; a free plan is available.

APIgency Hourly Earnings is a connector that connects Claude to Workday.Money, an earnings and client-work tool for freelancers and contractors who bill by the hour. Each client project has its own hourly rate, and earnings grow as work runs. Claude can manage the user's own client projects, start, pause and finish client work, use project rates, set billable limits, mark work as unpaid, and report earnings from completed records. It is not an employee attendance, workforce-management, HR, or payroll system. Finishing a task records completed work in Workday.Money, and that record never changes afterwards. If the project has a connected, writable Google Sheet, a copy is also sent there. Reports use the records in Workday.Money. The connector can create projects with a currency and later change their name, rate, or reporting period; archive and restore projects; cap billable time; mark work unpaid; move unfinished tasks between projects; and trim forgotten time. Trimming only reduces accumulated time on an unfinished task, and the task must be paused first. The assistant cannot sign in as the user, pay, change the plan, or delete the account or projects. It can delete an unfinished task together with its still-unrecorded time. Timers run on the server, not in the chat. Ending the conversation does not stop them; a task stops when it is paused or finished. The app is free for one active project. Using the connector requires Pro or an active 30-day trial. Pro includes unlimited projects for $5 a month. There is nothing to buy inside the assistant. Workday.Money is an independent product and is not affiliated with, sponsored by, or endorsed by Workday, Inc. WORKDAY® is a trademark of Workday, Inc.

The Apiiro Guardian connector lets Claude securely access your Apiiro Application Security Posture Management (ASPM) platform. Powered by Apiiro's Risk Graph and Deep Code Analysis, it surfaces prioritized application and supply‑chain risks with code‑to‑runtime context, enriches AI‑assisted development with security requirements ("secure prompt"), and supports pre‑commit remediation so risks are caught and fixed before they reach production. Use it to investigate findings, track remediation, and embed security guardrails directly into your development workflow.

APIMock lets Claude build and debug hosted mock APIs while it writes the frontend or tests that call them. Mocks are served over HTTPS at your workspace's own subdomain (https://<workspace>.apimock.dev), so your app, test suite and teammates can call them right away. What Claude can do with it: - Build mocks: create endpoints with fixed or templated responses, scaffold several stateful CRUD resources at once with typed fields and seed data, or import an OpenAPI/Swagger spec. - Return different responses from one route: response variants matched on headers, query, path parameters, cookies or body fields, or chosen by weight. simulate_request shows which response a request would get, without sending it. - Debug integrations: inspect_failed_requests shows what your client actually sent (unmatched routes, failed validation, errors), and heal_mock_endpoint fixes the mock to match. - Test failure paths: inject latency, error statuses such as 503 or 429 with Retry-After, and broken payloads, with rules that expire automatically. - Read the contract: each workspace's openapi.json and types.d.ts are available as MCP resources. Sign in with your APIMock account over OAuth. APIMock is free: 50,000 mock requests a month per workspace, up to 3 workspaces with team members, and 7 days of request history.

O apipj é uma API de consulta de CNPJ que traz para o Claude os dados cadastrais públicos de empresas brasileiras pelo número do CNPJ, a partir dos dados abertos da Receita Federal, atualizados todo mês. O que dá para fazer: Consultar um CNPJ: razão social, nome fantasia, situação cadastral e data, natureza jurídica, porte, capital social, data de abertura, endereço completo com código IBGE, telefone e e-mail cadastrados, CNAE principal e secundários, opção pelo Simples Nacional e MEI e o quadro de sócios e administradores (QSA). Validar um CNPJ (dígitos verificadores), inclusive o novo CNPJ alfanumérico, sem login e sem gastar cota. Ver o mês dos dados carregados. A partir de 28/10/2026: listar empresas ativas por atividade (CNAE) e município, sem dados de sócios. Exemplos: "Qual a situação cadastral do CNPJ 11.378.117/0001-20?", "Quem são os sócios dessa empresa?", "Essa empresa é MEI ou optante do Simples?", "Confira se estes 20 CNPJs da planilha estão ativos", "O CNPJ 12.ABC.345/01DE-35 é válido?". Como conectar: adicione o conector e entre com sua conta Google. A conta grátis não pede cartão e inclui 50.000 consultas por mês (10 por minuto); planos pagos aumentam os limites. Validação e status funcionam sem login. Os dados são públicos e não têm valor de certidão; a situação pode ter mudado depois da última carga mensal da Receita. CNPJ API: Brazilian company registry lookup with official open data from Receita Federal.

apMZoomAI connects your assistant to the Dongdaemun wholesale fashion market in Seoul, Korea, where stalls in many buildings supply clothing, shoes and bags to buyers worldwide. Ask in your own language (English, Chinese, Korean, Japanese, Vietnamese, Thai, Indonesian or Malay) and the assistant can: - search items listed by stalls, with photos and category; - show new arrivals from the last hours or days, by building and category; - find a stall by name or stall number, or list stalls by building and floor; - list the market buildings, with opening hours and addresses where known. Every result links to the item or stall on apMZoomAI, where buyers see prices and wholesale ordering details. The server is public and read-only. It never returns prices, stock levels or merchant contact details, and it cannot place orders. Only stalls currently in business are included. Item names and photos come from the stalls themselves. No account is needed. Operated by apM Hwashin Co., Ltd., Seoul.

Apogee connects Claude to read-only research tools over US federal and state government records. Coverage: federal and state bills with full text, versions, cosponsors, actions and roll-call votes; committee schedules and published hearing transcripts; CRS, GAO and CBO reports; Federal Register rules, dockets, comment periods and executive orders; Senate LDA lobbying filings by bill, organization and topic; FEC PAC contributions and independent expenditures for members of Congress and bill sponsors; federal awards and appropriations; members of Congress, state legislators and caucus rosters; and policy news and member press releases. Apogee keeps its own copy of these records and searches the text directly. Results link to the source where the source publishes a link: Congress.gov, GPO, the Federal Register, FEC, Senate LDA filings and state legislature sites. Search results are ranked by relevance and paged. When more matches exist than a page shows, the response says so. News and press releases are labeled as such and are not part of the official record. Limits: official hearing transcripts appear months after a hearing. State lobbying data covers only the states the coverage tool lists. Some analytics tools draw on the account usage allowance. Built for lobbyists, trade association staff, advocacy organizations and in-house policy teams that track legislation at the state and federal level.

Apollo is a B2B sales intelligence and sales engagement platform connected to Claude. Search Apollo's database for people by title, seniority, location, and time in role, and for companies by industry, size, technology stack, funding, hiring, and headcount growth. Find companies using a specific technology and the senior people who work there. Surface hiring, funding, and headcount-growth signals from company data and job postings, then save people as contacts and companies as accounts and organize them in lists. Enrich people with work emails and, where available, phone numbers. Revealing phone numbers uses additional credits. Enrich companies using their website domains to find industry, headcount, revenue, and funding details, one at a time or in batches. Keep your pipeline moving: create and search deals, log phone calls, and manage tasks. Draft personalized, multi-step email sequences and create them in Apollo for your review. Add contacts to existing sequences with your confirmation, and track sent emails, opens, and replies by sequence.

App Ratings Tracker records your App Store ratings, rating counts and app versions every day, so you can see how a release actually moved the numbers instead of guessing from today's snapshot. Through the connector you can, in plain language: Add an app by name and start tracking its public ratings See rating history by storefront, with the app version that was live each day Read customer reviews from App Store Connect, filtered by star rating, territory, or whether they still need a reply Draft a reply in your app's own tone, then queue it for approval Nothing is published to the App Store by the assistant. Drafted replies are saved to your dashboard at appratingstracker.com, where you read, edit and post them yourself. That approval step is deliberate: a public reply on your listing is permanent, and a person should be the one to send it. What you need: an App Ratings Tracker account, which you connect once via OAuth. Tracking public ratings needs nothing more than the app itself. Reading and replying to reviews additionally needs an App Store Connect API key, which you upload in Settings; it is encrypted at rest and never returned to a browser or an assistant. Works with any app on the App Store, across every storefront we track.

Attri connects your short links, your site's traffic and your Google Search Console data to what visitors actually did, so an assistant can answer the questions that usually take three tabs and a spreadsheet. Ask what brought visitors this month and which sources converted. Ask which search queries and landing pages gained or lost clicks, with the sessions and conversions behind each page, not just impressions. Ask what to fix first: Attri's posture review turns Search Console and on-site behaviour into ranked findings (title rewrites, pages to strengthen, cannibalised queries, drops to investigate), each with its evidence, the suggested action and the metric it will be measured by. Pull your short links and the clicks and conversions each one drove. Reads cover site overview, top pages, sources by channel (search, social, AI assistants, referral, direct, email), outbound clicks, Search Console queries and pages, the posture findings, links, conversions and goals. Two optional write tools, behind their own permissions, create a short link with UTM parameters and plan, complete or dismiss a finding. You sign in with your Attri account and choose which workspaces the connector may use and what it may do; a workspace admin can see and revoke connections. Search Console data needs the Attri workspace to have Google Search Console connected. Attri never reads your conversation, only the workspace data you enabled.

Bansaj is a genealogy platform for recording Nepali family lineages (वंशावली). With this connector Claude can find people in Bansaj by name, place or occupation, walk their family tree down to spouses and descendants, trace their parents and grandparents, and work out how two people are related. Names can be searched in Nepali or English script. The connector is read-only, returns only information already public on Bansaj, and needs no sign-in.

Cotera is where teams build AI agents on top of their own data and tools. This connector puts that workspace inside Claude: browse the agents you already have, build new ones, run them, and bring the output back into the conversation. Build agents. Describe what an agent should do in plain markdown and Claude creates it in your workspace, with headings, lists, and code blocks intact. Prompts can reference your connected tools, your documents, and your other agents, so a new agent delegates work the same way one built in the Cotera editor does. Find what to reference. Search the tools your workspace has connected — including whether each one is currently connected and ready to run — and search your documents, so an agent is wired to the real thing rather than to a guess. Run them, one input or many. Run an agent against a single input, or start up to 50 runs in parallel over a list: a batch of leads, tickets, transcripts, or rows. Runs execute in the background, and you collect the outputs as they finish. Read what you have. List every agent in the workspace, newest first, including ones your teammates built in the Cotera app, and read any agent's full prompt. Everything happens in the workspace you authenticate with. Agents created here are ordinary Cotera agents — your team can open, edit, schedule, and share them in the app afterwards. Nothing is destructive: creating an agent never modifies an existing one, and running an agent never changes it. Two things I deliberately kept in, since they're the details that make the listing trustworthy: the 50-run batch ceiling (the real MAX_BATCH_SIZE) and the read-only/non-destructive framing, which matches the tool annotations you ship.

Heep is the AI assistant restaurants use to answer guests on WhatsApp, Instagram, Messenger, email, SMS and Apple Messages, take reservations in their booking system (Zenchef, SevenRooms, TheFork, e-table), recover missed calls and bring guests back. This connector lets you run your Heep account directly from Claude. Get started in minutes • Create your Heep account during the connection, then set up your venue from its Google Maps listing. • Teach the AI: add facts to its memory (hours, terrace, allergens, policies), adjust its tone and instructions, add temporary instructions ("no terrace this weekend"). • Test it: send a sample guest message and read the AI's reply before going live. Run your restaurant from Claude • Read and search conversations, see which ones need your team, reply to a guest, pause the AI on a thread. • Look up reservations and guest profiles: visits, no-shows, upcoming bookings. • Get your numbers: messages handled, automation rate, revenue generated by bookings, ROI, missed calls. • Manage reviews from Google, TripAdvisor and TheFork: list them, draft replies in your venue's voice, publish to Google. • Configure reminders, review requests, missed-call follow-ups, WhatsApp template and campaign drafts. • Pick a plan and get a secure Stripe checkout link. Safe by design Everything runs with your own Heep permissions; Claude only sees your venues. Every change is logged in your activity log. Sensitive actions (messaging a guest, publishing a review reply, rewriting the AI's instructions, sending a campaign) always show a preview and wait for your explicit confirmation.

Hookie is a webhook platform: it captures anything sent to you — a provider's JSON, a plain HTML form, a scheduled job, a WebSocket feed — routes it into datasets with rules you define, and delivers signed, retried, replayable events onward. This server lets a coding agent do all of that on your behalf, from wherever you already work. Connect it and your agent can read your workspace (projects, datasets, records, routing rules, endpoints, destinations, recent deliveries and their response codes) and build in it (create endpoints and rules, point events at a destination, define multi-step AI workflows, mint ingest keys, replay a failed delivery). Asking "why did last night's delivery fail?" and having the fix applied is one conversation rather than a tab-switch. Nothing is installed and no key is pasted anywhere. The server is hosted at https://app.hookie.ai/mcp; your agent registers itself, you sign in once in the browser, and it holds a token it refreshes on its own — standard OAuth 2.1 with PKCE. A newly connected agent can only read until you widen it in the console, and revoking takes effect on its very next call. It acts as you, with your workspace role narrowed to the scopes you granted — whichever is more restrictive wins. Billing is unreachable at every scope, and no tool can delete a project or drop a dataset's records. Every action lands in your audit log with you as the actor and the agent's client id recorded beside it.

Manage your Omnara cloud agents via the Omnara MCP Server. Omnara exposes a stateful, open-source agent API so you can build: * Agents in your product's dashboard * Internal Slack Agents * Specialized agent workflows * Interactive CLI experiences The Omnara MCP server can allow you to: * Create and configure agents * Interact with agents * Create sandboxes * Manage model providers * Manage prompts, skills, and MCP servers

ReportGarden connects Claude to your agency's ReportGarden account, so you can ask about your clients' marketing data in plain language. ReportGarden pulls data from over 100 advertising, analytics, SEO, and ecommerce platforms into one place, including Google Ads, Meta, LinkedIn, TikTok, Amazon Ads, Google Analytics 4, and Shopify. With this connector, Claude can: - Answer performance questions. Query live account data by any metric, dimension, date range, or filter. For example: "How did spend and ROAS change month over month for Acme's Google and Meta campaigns?" - Pull numbers from several ad accounts and integrations to compare channels or spot trends and anomalies. - Find what you need. Look up clients, linked accounts, campaigns, budgets, custom metrics, and the metrics and dimensions each integration offers. - Explore your reports. Search your reports and templates, see what's in a report, and pull the data behind any widget. Sign-in uses OAuth, and Claude can only see what your ReportGarden user account can. A ReportGarden account is required.

Reright is a review step for text that agents write and people read. An agent calls submit_for_review with a commit message, PR description, email or client message, plus the context a reviewer needs. You open a web page or a push notification, edit the text and approve or reject it. The agent then calls wait_for_review and gets back either the approved text, which it must use exactly, or a rejection, which tells it to stop. In Claude Code, the reright plugin adds hooks that block commits, PR and issue text and emails until the exact text was approved. Other MCP clients can use the two tools directly, but they cannot enforce the check. Sign in with OAuth. The free plan covers 100 messages a month. Text is encrypted at rest and in transit and deleted after the retention you choose (1 day on Free, 1 to 90 days on paid plans) and from every backup within 48 hours after that.

Siftree is the fastest way to analyze customer videos, documents, images, audio, and text at scale. Tap into a marketplace of over 1,000 public sources like TikTok, Amazon Reviews, AI Search Results, US Congressional Hearings and more. Kick-off colossal AI workloads and let Siftree manage storage, compute, compliance, and governance for you.

Search AppAgg for apps and games by name or developer, then retrieve concise summaries with links to full AppAgg pages. Full descriptions, screenshots, price history, release details, and downloads remain on AppAgg.

Appbot has analysed App Store and Google Play reviews for over a decade, across every category and language. The AI behind this connector is trained on our corpus of 400M+ real reviews. Connect your App Store and Google Play apps, and Claude queries your review data directly instead of working from raw text: What is driving negative reviews this month? Which feature requests have grown over the last 90 days? Why are iOS ratings lower than Android? How has sentiment shifted since our latest release? How does our sentiment compare to a competitor's? Competitor apps work the same way, so you can benchmark sentiment and feature gaps against anyone in your category. Each team member connects with their own Appbot credentials, so Claude only sees what that person can already see in Appbot. MCP access is included on Appbot's Large and Premium plans.