Explore Claude connectors.

fAlpha is a pre-trade validation layer: machine-learning-generated up/down signals with the explanation behind them, not just a black-box call, so you validate ideas before you commit capital. Query the signal with point-in-time history and risk-adjusted track record, screen the market by live signal, pull daily news sentiment, compare sell-side analyst calls against the model's own accuracy, and pull SEC filings, macro data, and fundamentals all with disclosed data sources– all from one connector.

AgentAvow tells you whether a tool is safe for your AI agent to connect to, before it connects. Point it at a GitHub repo, an npm, PyPI, crates, Docker or Hugging Face package, or a live MCP server. Every result leads with one of three answers and the reason behind it: - Safe to connect: nothing blocking was found. - Review before you connect: a high-severity finding, a published advisory for that version, a deprecated package, or a low trust score. - Do not connect: a critical finding, a known-malicious dependency, or a planted test credential leaving our sandbox. "Certified" marks tools that also passed every provenance check. Under the answer sit two scores: a 0 to 100 trust score built from the findings, and an adoption score built from real usage (downloads, stars, installs). Each finding says where it is and how to fix it. Packages and MCP servers can also be run in a locked-down sandbox to see what they actually do: network calls, file writes, and whether planted credentials stay put. Every result is signed (Ed25519) and can be recomputed offline against our public keys, so you can check it without trusting us. Read-only. No account. Ask it to check a tool before you install or connect it.

AgentCollect's public MCP connector lets Claude query your accounts-receivable recovery data. It exposes one read-only tool, ask-ar-analyst, which answers natural-language questions about outstanding invoices, collection status, disputes, and recovery performance. Access is scoped to your own account via OAuth (dynamic client registration, PKCE S256, scope mcp:read) and no data is modified.

Fivetran's Agent Context MCP provides AI agents the full breadth of business context needed to answer questions and take action, reliably. Built on Fivetran's open context layer, it infers and curates context from every enterprise systems — including data sources, schemas, concepts, knowledge bases, and analytical metadata — and exposes it through the Model Context Protocol. Agent Context MCP powers natural-language search, data-aware assistants, coding workflows, and custom AI applications. Its standard MCP interface makes it easy to bring rich, trusted, and comprehensive business context into the AI tools teams already use, including Claude, ChatGPT, coding agents, and custom applications.

AgentCron is your team's Collaborative OS. Plan. Execute. Measure. Bring your team's work into the conversation. Find projects and tasks, summarize progress, create or update task cards, organize Drive files and folders, and read or send workspace messages. Turn a conversation into work your team can see and act on. Connect your AgentCron account with OAuth, choose a workspace and approve access. The connector follows your existing workspace and resource permissions. Disconnect it in AgentCron Settings > Connected apps. An AgentCron account and workspace are required. A free plan is available. This connector works with AgentCron's projects, boards, cards, files and chat; it does not process payments or replace the full AgentCron interface.

Agentel connects Claude to public opportunities, capabilities and evidence in the Agentel network. Discover public Missions, search for Skills or Agents, inspect eligible items and read participation guidance. Check your bound Agent identity and allowance before metered reads. Six read-only tools cannot apply, publish, purchase, install or execute a Skill; authorized writes use the separate Agentel SDK/API. Sign in to Agentel through OAuth and choose an Agent reading credential you control. Accepted reading attempts, including some empty results, may consume INFO and post allowances. Reading a Mission does not earn rewards. Claude supplies the model; this connector does not invoke Agentel's MiniMax account.

Agent Grid connects Claude to a collaborative workspace where AI agents and humans can create, upload, view, update, and share interactive artifacts such as documents, designs, prototypes, front-end experiences, and presentations. Related artifacts can be grouped into shared workspaces, while human ownership, agent authentication, granular permissions, and team-based entitlements keep collaboration governed and secure. It is like Google Drive but built for Agents and Humans.

Agentibus runs AI sales agents for coaches, course creators and education businesses. The agent replies to every lead on Instagram DMs, WhatsApp, Messenger, web chat and email, qualifies them against your criteria and books sales calls into your calendar. Across our customers it has booked more than 7,000 sales calls and handled more than 70,000 leads. With this connector you run your agent from Claude: - Pipeline: list your leads, read any conversation, see funnel metrics, bookings and time saved. - Approvals: in hybrid mode every reply waits for you. Review the drafts, then approve or discard them. Approving a reply, discarding a draft and cancelling follow-ups first return a preview of the exact action and run only after you confirm. - Control: pause or resume an agent, take over a single lead by hand, cancel the follow-ups scheduled for a lead. - Knowledge: add a web page or a text (FAQ, price list) to what your agent knows, and test the agent in a sandbox before it talks to real leads. The connector needs an Agentibus account. You sign in with OAuth, act only on your own workspace, and every action is recorded under your name. New to Agentibus? Your first 10 booked sales calls are free: start at agentibus.ai.

Agentic Atlas helps you design and review agent systems: skills, subagents, plugins, hooks, and multi-agent workflows. Connect it, then ask Claude a design question in plain language. Claude looks up the patterns that fit your situation and answers from their published rulings, which you can check on agentic-atlas.dev. Reach for it while a design decision is still open: splitting a system into skills, subagents, or tools; deciding what a subagent carries in and hands back; keeping context lean in a long-running workflow; placing a verification step so a wrong answer gets caught; or reviewing a skill or plugin you have already written. Each pattern states a ruling, a worked example, and its links to related patterns. Every ruling links to the decision record that settled it, and every entry carries a maturity label, so you can tell field-tested advice from advice still earning its evidence. A glossary defines the working vocabulary, and a curated eight-step tour is the place to start. The connector is public and read-only. There is no sign-in, account, or payment, and it never writes anything. Setup and the full tool reference: https://agentic-atlas.dev/connect

AgenticPM - Deal Scout is a deal-sourcing connector for private equity investors, private-markets funds and the advisers who work with them. It turns a plain-language sourcing request into a structured, sourced result. Describe the companies you're looking for, or name seed companies you want lookalikes of, and get a screened target list. The list is built from published rankings, trade directories, association rosters and other list pages rather than from search rankings. Each company is sorted into targets, close calls or excluded against your criteria, with the reason shown. Enrich named companies with ownership, parent firm, revenue, EBITDA, employees, headquarters, customers and competitors. It also rates deal signals such as succession pressure, sale readiness and acquisition activity, turning a list of names into something screenable. Extract specific facts from companies' own websites at scale, using standard fields or a question set you supply. Track recent news and deal announcements for a company, sector or event. Run wider private-markets web research to verify ownership, map PE portfolios and follow sector deal activity. Every value comes with its sources and a confidence rating. Where sources disagree, every value is shown, and estimated figures are labelled as estimates. Deal Scout researches companies, not people: it does not research, name or profile individuals. Five tools cover this funnel end to end: target discovery, company profiling and deal signals, company website data, private-markets web research, and news tracking.

Agenticsis Visibility Monitor reports how your company currently appears in AI answers and organic search. Connect it and ask Claude: How often do ChatGPT, Perplexity, Claude, and Google AI Overviews cite my brand for my tracked queries, and with what sentiment? What is my Share of Voice versus each competitor this month, and how is it trending? Where do my keywords rank, what do my backlink and traffic profiles look like, and which of the 54 checks in my AEO/SEO scorecard does my site pass or fail? What did the full-site technical crawl find, which entity signals (Wikidata, Crunchbase, schema consistency) is my Knowledge Panel readiness missing, and what does this week's market intelligence (fresh industry news, Perplexity research brief, real search volumes) say about my space? Every number is measured monthly by the Agenticsis pipeline from data sources Claude cannot query alone (DataForSEO, Firecrawl, Perplexity) and is scoped to your own company. Read-only measurement and reporting. Requires an active Agenticsis Visibility account.

The Agentio connector brings your Agentio reporting into Claude. Ask for your Creator advertising performance in plain language and get the numbers back with the definitions behind them, so a question like "what did we pay per conversion last quarter?" is answered rather than assembled from a report. Ask which Creators drove the most engagement this quarter, how cost per acquisition changed month over month, how a campaign's reach is holding up week to week, or how a 30-day view is counted.

AgentiSend is a transactional email API for AI agents: send through verified domains, with a budget each sender cannot exceed and a kill switch a person can flip. The hosted MCP server at https://api.agentisend.com/mcp exposes 81 tools (43 on tools/list; the rest via list_more_tools). A send is one message or a batch of up to 500 items with per-item results. preflight_email runs every gate a real send runs, sends nothing, and costs nothing. Every 4xx returns code, message, and fix. There is no tool for unsolicited mail. Auth is a bearer API key or OAuth 2.1 with dynamic client registration. Starter prompts: "Send this receipt to the customer." "Preflight this send and tell me what would stop it." "What can I spend today?"

Agently Brain gives any MCP client a company brain: a temporal knowledge graph continuously built from everything a company connects to Agently — uploads, Slack, Gmail, HubSpot, Linear, GitHub, and 25+ other sources. Unlike per-session memory servers, the brain is shared across every agent and tool a team points at it, and its facts carry temporal validity, so agents can distinguish current truth from history. Six tools: search (hybrid semantic + keyword, results carry entity UUIDs), find_entity, get_neighbors (graph walking), get_schema, recent_episodes, and remember (write-scoped). Auth: workspace API keys or OAuth 2.1 with dynamic client registration. Every credential is pinned to one workspace server-side.

AgentMail gives AI agents their own email inboxes. Through this connector, an agent (or you, in Claude) can spin up a dedicated inbox on the fly, then send, receive, reply to, and forward email, with full thread context, drafts, and attachments, entirely through tool calls. Unlike traditional email APIs built for one-way notifications, AgentMail is built for two-way conversations: an agent can read an incoming thread, understand it, and respond in context, just like a person would. This makes email a first-class communication and identity channel for agents, letting them sign up for services, coordinate with people, and talk to other agents. Key capabilities exposed over MCP: • Inboxes: create, list, get, update, and delete agent inboxes • Messages: send, reply, forward, list, and update • Threads: list, read, label, and delete full conversation threads • Search: full-text search across threads and messages • Drafts: create, list, read, update, send (with scheduling), and delete • Attachments: retrieve attachments by ID, with text extraction for PDF/DOCX Connect in seconds via OAuth using your AgentMail console identity. No API key required in Claude. Sign up free at console.agentmail.to.

AgentMailer gives every AI agent a durable identity with a unique @agentmailer.ai handle. Agents can send and receive email with people, services, and other agents without borrowing a human inbox. Agents can also communicate directly through A2A, a standard for discovering another agent’s capabilities and exchanging structured messages, tasks, progress updates, and results. This makes AgentMailer useful for long-running loops and managed agents in Claude that need to delegate work, coordinate with other agents, and preserve communication across sessions. Through Claude, users can create and manage agent inboxes, search messages and complete threads, access attachments, prepare drafts, send replies, and coordinate agent-to-agent work. Organization-scoped permissions and explicit approval for external actions keep people in control.

Agentman is the team library for Agent Skills — the open SKILL.md format behind Claude Skills. Anyone can share a skill. Agentman is for teams whose skills carry real expertise — legal playbooks, clinical procedures, deal frameworks — where a bad edit costs money and "who changed this?" needs an answer. What Claude can do through this connector: - Search your library by keyword or in plain language ("how do we screen a deal?") - Open and apply any skill you have access to - Read bundled references, templates, and scripts - Create, update, version, and publish skills from chat - Browse skill sets, clone public skills, import/export .skill packages - Say "show my skills" to open the Skills Explorer — a clickable, filterable view of your library, right in the chat What your team gets underneath: - Share skills people can run but not open — clients and teammates use the skill; the know-how inside stays yours, and every use is logged - Changes go live only after review, and what's approved is exactly what was reviewed - You decide who can use, read, edit, or admin each skill — per skill or across the workspace - Full version history: compare any two versions, add notes, restore in one click - Always know who used what, when — per skill and across the workspace - One workspace per client or matter, so nothing leaks between them Built for legal, private equity, healthcare, and other teams that can't afford to get this wrong. One library, every Claude Surface: the same skills run in Claude.ai, Claude Cowork, Claude Code, and Claude Managed Agents via OAuth 2.1 — per-member, revocable connections. 100+ production-ready skills across 23 business functions, ready to clone and customize. Free to start.

AgentMart Site Check reads the public setup of a website, the way any visitor or AI agent sees it from the outside. Give it a domain such as example.com. It checks HTTPS and security headers, certificate dates, email sender records (SPF and DMARC), whether privacy, terms and contact pages can be found, and files for AI agents (llms.txt, robots.txt and MCP). Each item comes back as pass, warn, fail or not checked, with plain steps to fix it. How it reads: at most 24 requests to the site per check, one of them over plain http to see whether the site moves visitors to HTTPS, plus public DNS questions. If the site names an MCP server, the check sends it the standard MCP opening messages, asks for its list of tools and ends the session. It follows the site's robots.txt rules for AgentMart-Check. It never logs in, fills in forms or changes anything. What it returns: our own fixed wording and numbers. No text from the checked site is copied into a result. A result is kept for 7 days and can be shown again by its check_id. A repeat for the same domain within 15 minutes gives the same result, unless you ask for a fresh check. Limits: a result describes that moment only. It is not a security audit, and it says nothing about the company's servers, code, staff or how it handles data. No account and no sign-in. Site owners can stop checks of their domain with a DNS TXT record agentmart-check=no on _agentmart-check.<their domain>, or by writing to us. Made by Ciphyr Tech OÜ (Estonia), the company that runs AgentMart.

AgentMash brings everything you saved on social and video sites into one searchable library. This connector lets Claude read that library. Ask Claude about something you remember saving. It searches your library, opens the full saves, and answers with a link to each source. It covers posts, threads, carousels, short and long videos (with transcripts where available), comments and articles. What Claude can do: - Search your saves by keyword, and by meaning on plans that include semantic search - Filter by source, type of save, author, collection or how recently it was saved - Open one save in full: complete text, transcript, links, media and stats - List your recent saves, your connected sources and your collections The connector is read-only. Claude can search and read your saves. It can't post, edit, label or delete anything, and it only sees the library of the account you sign in with. You sign in to AgentMash and approve read-only access on the consent screen. You can disconnect Claude at any time in AgentMash under Settings, Connected apps. Before you connect, you need an AgentMash account with at least one source synced through the AgentMash browser extension. The Free plan includes 50 agent calls a month. Pro includes unlimited calls.

Agentmatica is the AI shopping assistant that turns shopper questions into orders on Shopify, WooCommerce and other stores. With this connector you set it up and train it by chatting with Claude instead of filling in forms. Give Claude your store's address and it creates an agent for it, starts reading the site, and gives you the code to install it (on a connected Shopify store, the theme app embed steps instead). Ask later whether it has finished, and it tells you how many pages it has read and whether the chat has shown up on your site. Tell Claude how you want shoppers handled, for example which product to suggest alongside another, what to say about returns, or how many questions to ask at once. We show Claude your agent's existing rules so it can explain how the agent behaves today and avoid duplicates. Claude turns your instructions into rules in our format and checks them for gaps and clashes. It saves them once you agree, and saved rules go live on your agent straight away. You sign in with your Agentmatica account, or create one on the sign-in page, and can only reach agents you are allowed to edit. Agents you create count toward your plan. The connector has no access to chats, customers, orders or billing. Disconnect any time from Account > Connected assistants in Agentmatica.

AgentPricing Connector è il collegamento tra Claude e AgentPricing per gli agenti immobiliari. Da chat puoi preparare valutazioni, confrontare venduti e annunci, verificare un prezzo richiesto, analizzare il mercato locale e riaprire i report esistenti. Serve un account AgentPricing autenticato.

AgentSandboxes lists cloud sandbox providers for AI agents. Each listing shows isolation, cold start, persistence, GPU support, pricing and free tier from the vendor's own docs, with every fact linked to its source. Browse by category, compare providers side by side, or query listings through the free REST API and MCP endpoint with no key needed.

AgentTrust looks up an AI agent by the exact URL you are about to call and returns what AgentTrust has observed about it: the agent's current health status, whether the endpoint's ownership has been verified, its reliability score from AgentTrust's own monitoring history (when one is available), and a trust decision with a recommendation, a confidence level and the reasons behind it. The connector has one tool, check_agent_trust. It is read-only and needs no AgentTrust account or API key. It never contacts the endpoint being checked: it only reads the monitoring data AgentTrust has already collected about public agents, including agents registered by their owners and MCP servers listed in the official MCP Registry. If AgentTrust has not observed a URL, the result says so ({ matched: false }) rather than guessing. A trust decision is a signal derived from AgentTrust's observations, not a certification or a guarantee that an agent is safe, so treat it as one input alongside your own judgment. URLs must be absolute http:// or https:// addresses, and anonymous checks are rate-limited. Each check records minimal usage data (no raw IP addresses or full URLs); see the privacy policy for details.

AgentUI Apps lets Claude create and manage data apps in your AgentUI workspace. A data app is a hosted database with tables and records, without a website. What you can do: - Create an app and define its tables and fields (text, numbers, dates, emails, lists and more). - Change a table later: add, rename or remove fields. Existing records are migrated for you. - Find, count, create, update and delete records, one at a time or up to 200 per call. - Write programs: short steps of code that read and write your app's data. Run them on demand, on a schedule, from a webhook, or when an email or WhatsApp message arrives. Each run returns step-by-step logs. Every app you open also comes with its own MCP connector URL, so you can share just that app's data with another tool. Deleting a table or removing a field needs an explicit confirmation. Read-only tools are marked so Claude can use them without asking. You need an AgentUI account to connect. Sign in with OAuth; Claude only sees the apps in your workspace.