Explore Claude connectors.

Bencher is a continuous benchmarking platform that helps systems software engineers catch performance regressions before they merge. How Bencher Works: 1. Run: Run your benchmarks locally or in CI using the exact same bare metal runners and your favorite benchmarking tools. The `bencher` CLI orchestrates running your benchmarks on bare metal and stores the results. 2. Track: Track the results of your benchmarks over time. Monitor, query, and graph the results using the Bencher web console based on the source branch, testbed, benchmark, and measure. 3. Catch: Catch performance regressions locally or in CI using the exact same bare metal hardware. Bencher uses state of the art, customizable analytics to detect performance regressions before they merge. For the same reasons that unit tests are run to prevent feature regressions, benchmarks should be run with Bencher to prevent performance regressions. Performance bugs are bugs!

Ask Claude for a business website, landing page, portfolio or event page. Benchlight saves a private draft to your account and shows phone and desktop preview pictures. If you already have a website, confirm it is yours and Claude can read its public pages and eligible pictures to keep the real business details. Before moving a domain, check its current website and email records without changing them. After you allow free publishing and say yes to a publication in the chat, Claude can put the draft online as your one free benchlight.page site, with a small Benchlight credit. Private drafts expire 30 days after the last change; up to ten recent versions are kept. You can also publish from Benchlight or use hosting on your own domain. Claude can read recorded daily visitor stats for your sites. Benchlight hosts static HTML, CSS, JavaScript, images and fonts. The connector does not buy domains, start paid hosting or change DNS settings.

Search for trusted causes to donate to or volunteer with from Benevity's database of more than 2.3 million verified nonprofits.

beschlussreif.de ist das Wissens- und Werkzeugportal für selbstverwaltete Wohnungseigentümergemeinschaften (WEG). Mit diesem Connector greift Claude auf die veröffentlichten Ratgeber des Portals zu und rechnet zwei gesetzliche Fristen nach. Was Claude damit kann: - Die Ratgeber durchsuchen – zu Eigentümerversammlung, Beschlüssen, Umlaufbeschluss, Hausgeld, Wirtschaftsplan, Jahresabrechnung, Verwalterwechsel und Vermietung. - Einen Ratgeber vollständig lesen: Kurzantwort, Schritt-für-Schritt-Anleitung, Checkliste, Rechtsgrundlagen und Stand. - Prüfen, ob die Einladung zur Eigentümerversammlung die Drei-Wochen-Frist einhält (§ 24 Abs. 4 WEG). - Die Fristen für eine Anfechtungsklage berechnen – einen Monat für die Klage, zwei Monate für die Begründung (§ 45 WEG), mit den Feiertagen Ihres Bundeslands. Jede Antwort nennt den Stand des Ratgebers und die Adresse, unter der Sie ihn auf beschlussreif.de nachlesen können. Ohne Anmeldung und ohne Konto. Der Connector liest nur und hat keinen Zugriff auf Daten einzelner Gemeinschaften oder Personen. Suchbegriffe und Daten werden beantwortet, aber weder gespeichert noch protokolliert. Die Inhalte sind allgemeine Informationen und ersetzen keine Rechtsberatung im Einzelfall.

BESTPRED gives Claude USDA’s best prediction method, the one used for official dairy lactation records since 1998, so it can work out lactation records from test days instead of estimating them. Ask for a herd’s 305-day records and Claude sends the test days (from a DHI or DairyComp export) to BESTPRED, which returns, for every lactation: projected 305-day and 365-day milk, fat and protein; mature-equivalent yields when the cow’s age and state are known; reliability; persistency; and a summary by parity with the least persistent cows. It can also project one cow’s record, draw her expected daily milk to day 305 to compare with what she’s giving, and give USDA’s mature-equivalent factors. The calculations are a Python port of USDA-AGIL’s BESTPRED 2.0rc7, checked against USDA’s own program (every output agrees to 1e-9). Test days sent to the service are used only for that answer and are not stored. To keep data on your own computer, install it locally instead: uvx bestpred-mcp. Not affiliated with or endorsed by USDA. Open source (Apache-2.0): github.com/padge01/bestpred-python

Besty AI searches real inventory from professional vacation rental managers, shows day-by-day rate calendars and reviews, produces live quotes with full fee and tax breakdowns, and hands you a checkout link on the property manager's own website with your dates and guests pre-filled. Booking direct means no third-party channel markup. Prices come straight from each property's management system, cached calendars say exactly when they were refreshed, and final totals are always quoted live.

BETA — Climbing gym management data access. Provides read-only administrative access to a climbing gym's operational data via the BETA (sendmoregetbeta.com) platform. Supports querying gyms you administer, searching climber/member records, retrieving membership passes and punch cards, listing products for sale, viewing check-in session logs, and reviewing transaction history. Useful for gym administrators who need to look up members, audit check-in activity, track pass usage, or review sales and revenue data.

betadomain is a German domain registrar focused on transparency: register, transfer and manage domains across ~490 TLDs at fair, fully disclosed annual prices — no first-year lock-in pricing, no hidden surcharges, no forced auto-renewal (switch it off anytime, with reminders), and a free, instant auth code when you move a domain away. This connector lets an AI assistant check availability and prices, look up non-personal WHOIS/RDAP data, list your domains, check order status, and start an order. Read tools need no sign-in; account and ordering tools use OAuth. Buying always finishes with a human checkout (payment and, where required, identity verification) — nothing is charged automatically.

BetBuilders AI brings model-based football and Swedish ice hockey analysis into Claude. Ask about a match to get the model's win, draw and loss probabilities (hockey: after 60 minutes and including overtime), expected goals, the most likely scores and the distribution of total goals. Read the match analysis with recent form, each team's xG trend over the last ten games, table position, KPIs with league ranks and key players. Look up a team's season: points versus expected points, xG, xGA and xGD per match with league ranks, streaks and recent results. Run what-if scenarios in the Monte Carlo match simulator – a key player missing, a different goalie, a stronger or weaker team or a neutral venue – and compare them with the baseline. Ask how likely several outcomes in the same match are together, with correlation taken into account. League tables include expected goals and expected points. Coverage: 40 football leagues and cups, including the Premier League, La Liga, Bundesliga, Serie A, Ligue 1, Champions League, Allsvenskan and Eliteserien, plus SHL and HockeyAllsvenskan. All tools are read-only and need no account. Every answer links to the full match or team page on betbuilders.ai. Probabilities are model estimates, not guarantees. BetBuilders AI is built and operated by PlaymakerAI.

Better Angles is a management platform for coaching and education businesses: lead pipeline, sales calls, payments, team commissions and an online school. This connector lets you ask Claude about that data in plain language. Claude can read your leads and pipeline stages, sales calls and their outcomes, payments received and installments still owed, monthly team commissions, revenue by currency (unified to USD), leads and sales by source (UTM, Instagram, Meta Ads), setter and closer performance, the top pains, objections and fears heard on won and lost calls, the health of your online school, and any KPI on your platform with its target and status. Access is read-only: Claude cannot create, change or delete anything. You choose the platform when you connect, and Claude sees exactly what your user sees when you sign in yourself: only the sections your role can open, and only your own records if your role is limited to them.

When Claude builds you an app, a website or a page, it often looks generic and misses how people will use it. Better Design is a design harness that helps Claude get both right. Pick a look. Describe your product, and Better Design shows you design systems that fit it, such as calm and trustworthy for a finance app. A design system is the set of colors, fonts, buttons and cards that makes a product look like one brand. You choose one, and Claude builds with it. Make your own. Describe the feel you want, or share your website, and Better Design creates a custom design system in a few minutes. Get the experience right. Better Design tells Claude how sign-up, onboarding, forms, search, checkout and error messages should work, and what each screen should say. Check the result. Better Design reviews what Claude built for the problems people notice: text that is hard to read, buttons that are hard to find, screens that ask for too much at once, and words only insiders understand.

Connect BetterBugs to Claude and bring complete bug reports directly into your AI-powered debugging workflow. The BetterBugs MCP (Model Context Protocol) server allows Claude to access BetterBugs bug reports, sessions, recordings, and workspace data using shared report links and MCP-enabled tools. Simply provide a BetterBugs report link along with your prompt, and Claude can retrieve the full debugging context, including screenshots, screen recordings, console logs, network requests, user events, environment details, AI summaries, and reproduction steps. With access to this rich diagnostic information, Claude can analyze issues, identify likely root causes, suggest fixes, explain technical findings, generate debugging plans, and help developers resolve problems faster. Beyond bug analysis, Claude can interact with BetterBugs resources to search and retrieve reports, review sessions and recordings, inspect project data, and perform supported BetterBugs operations directly through the MCP integration—creating a seamless bridge between bug reporting and AI-assisted development.

betterimage.io renders social cards for links: Open Graph and link-preview images (1200x630), YouTube thumbnails, Instagram squares, stories and Pinterest pins. Give Claude a page URL and it returns a card on a design picked to fit the page; browse 100 starter designs, preview any of them with your own text, check how a link previews on X, Facebook, LinkedIn, Slack and Discord, and get a ready-to-paste head block with Open Graph and Twitter tags. Sign in to save designs as templates, render clean images and get signed og:image URLs that produce a card per page with no hosting on your side. Every image is rendered from a designed template by a deterministic renderer: the text changes, the layout, fonts and colours stay on-brand. No generative image model is involved.

Bettership is a personal AI workspace for ambitious professionals leveling up with online learning. You bring the courses, podcasts, videos, articles, books and notes you chose, and Bettership helps turn them into ability you can use at work. This connector puts that learning inside Claude. Your Library. Ask Claude about anything you've added to Bettership. It can find a resource, read its original text, pull the passages that answer your question, and look up the ideas and methods your resources cover. Results carry each resource's title and link, so Claude can point you back to where an idea came from. What you've made. In Bettership, Maker turns what you've learned into prompts, context packs and skills. Publish one and Claude can find it, load it and follow it, without a file to install. When you publish a new version, Claude uses that one next time. What Claude can see. When you connect, Bettership shows the two things Claude is asking to read: your Library and your published items. Claude can read them and nothing more. It can't add, change or delete anything in Bettership. Your practice history, your conversations with your Bettership agents and your learner profile stay in Bettership. You can disconnect at any time in Bettership Settings. Requires a Bettership account on the Plus or Max plan. Sign in with an emailed code or with Google.

bFocus is a service desk and project management platform with built-in application monitoring. This connector brings that work into Claude. Monitoring: see the errors captured in your applications, grouped by failure, with stack traces, releases, affected customers and the linked task. Resolve, ignore or merge errors, open a task for an error, or request an automated investigation that reads the linked repository. Support: list and read tickets with their full conversation, add internal notes, and update status, priority or assignee using the same transitions as the app. With a separate permission, reply to the customer. Projects: list and read tasks, comment, update them, link a task to a ticket and create draft tasks; review projects with their risk radar, owners and open tasks. Release notes: list the changelog of every product and draft new release notes. Publishing stays with the team in the app. Several companies: if your email belongs to more than one bFocus company, connect them all at once, or add one connector per company with https://api.bfocus.com.br/mcp/<company>. Built to be safe to hand to an AI: - Every action runs on behalf of the person who connected and is recorded in the bFocus history under their name. - Replying to customers needs its own "Reply to customers" permission, unchecked on the consent screen; without it, ticket comments are always internal notes. The connector never opens tickets and never deletes anything. Tasks created by Claude start as drafts. - Content marked as confidential is never returned. - Tools only appear when your company has the corresponding module and you granted the permission on the consent screen. Reading is pre-selected; writing is your choice. Included workflows (prompts): a Monitoring round that triages new errors and regressions, fixing an error with the code open in your editor, ticket triage, and a weekly project health report. Requires a bFocus account with the admin or manager role.

Give your AI assistant exact spelling in all 22 scheduled languages of India. Bhasini's MCP server transliterates romanised text, converts between Indian scripts, and looks up keyboard maps with the same engine as this website. Connect to the hosted server with one URL, or run it on your own computer.

BIA Edge is a retrieval layer that gives AI agents citation-grade access to the U.S. immigration-law corpus. It covers published BIA precedent plus roughly 248,000 unpublished BIA decisions and 167,000 AAO non-precedent decisions, federal circuit, district, and Supreme Court immigration cases, Attorney General and OCAHO decisions, the INA and Title 8 and related CFR, the USCIS Policy Manual, FAM and agency guidance, state and federal criminal codes for crimmigration screening, visa bulletins, and EOIR statistics. Its read-only tools cover exact citation lookup, case and statute search, case-validity and treatment signals, unpublished and immigrant-favorable decision search, doctrine and related-precedent context, pinpoint source-text passages, and state and federal criminal-code screening. BIA Edge is a corpus-and-retrieval layer designed for the agent to verify source text and synthesize the work product; it does not produce final legal conclusions on its own. Built for immigration attorneys and the AI tools they already use. Access requires a BIA Edge account (request at https://biaedge.com/mcp-access); authentication is OAuth or an issued API key. The corpus is updated continuously, and freshness is reported through a corpus-status tool.

Connect Claude to Bible Atlas, a structured, cross-referenced map of Scripture: people, places, events, family trees, and original-language words, all linked to the verses they come from. What you can do - Look up verses. Fetch a verse or a range with its surrounding context. - Search by name or by meaning. Find an entity directly, or describe an idea in plain words and get semantically related results. - Explore people and places. Pull up an entity's profile, its family tree, and its connections to other people, places, and events. Search places on the atlas. - Check the original languages. Look up Hebrew and Greek lexicon entries. - Browse the datasets. List the collections behind the atlas, or call any other public API endpoint directly. Good for - Bible study and sermon prep - Tracing genealogies and relationships - Answering "where and who was this?" questions with sources - Grounding Claude's answers in actual verses instead of memory Details - Read-only. It can't change anything and only reads from Bible Atlas. - No account needed. It uses the public API.

Connect Claude to Bicycle AI, the agentic analytics platform for revenue-critical data. Ask questions about your business in plain language and get answers backed by your data. Browse your semantic models, look up metric definitions, and compare results across dimensions and time periods. Ask Claude to build a Bicycle AI data app from a prompt. Describe what you want to track, and it drafts the app, previews and validates its queries against your data, and helps you publish it for your team. Build executive briefings with trends, segment comparisons, and recommended next steps; include dollars at stake and progress against goals when the required metrics are available. Refine apps in conversation, compare versions, manage sharing, and schedule PDF or PNG snapshots of saved app views by email. For published apps with chat enabled, ask follow-up questions and generate summaries. Your connection is limited to the workspace you authorize and the permissions allowed by your role. Viewers can explore accessible data and read apps. Creating, editing, publishing, and sharing apps require the corresponding permissions.

Bid Intelligence connects Claude to Vietnam's National E-Procurement System (muasamcong.mpi.gov.vn), where public bodies publish tenders and award results. Ask in any language; Claude translates and summarises the Vietnamese data for you. What you can do: - Find tenders by keyword, goods, sector, province, procuring entity, budget or deadline: online price offers (CGTTRG), open bidding (DTRR), competitive offers (CHCT) and requests for quotation. - See tenders before they are announced: investment projects and contractor selection plans since July 2026, with each planned package's expected month, budget, method and funding source. - Read a tender in full: data sheet, evaluation criteria, technical specifications (Chapter V), attachments, clarifications, complaints, amendments, cancellations and extensions. - Price research: winning and offered unit prices for comparable goods, and a unit-price analysis for every item in a package. - Competition forecast: whether a tender is genuinely competitive, who is likely to bid and why, the price most likely to be lowest, and after bid opening each bidder's chance to win, with accuracy measured on recent results. - Profiles: a contractor's history and win rate, a procuring entity's purchases and usual winners, and partners with matching experience. - Market statistics by sector, province and period. - With a company account: your own bid results, watched packages and recommended tenders. All tools are read-only. The connector cannot submit bids or change your account, and it never sees your conversation, only the parameters Claude sends to each tool. Data is collected automatically from the official portal, usually within a few hours to one day. Sign in at dauthau.anovatech.vn; new companies get a free 14-day trial, and company reports and unlimited forecasts require a paid plan. Built by Anovatech (Vietnam). Support: info@anovatech.vn.

Search and compare more than 29,000 GAO bid protest decisions, from 1925 to 2026, without leaving the conversation. Describe an issue or fact pattern in plain language and get the closest decisions, each with its B-number, date, agency, outcome and a short summary. Open any decision by B-number for its issues, reasoning, why GAO ruled the way it did, key points and a link to the decision on gao.gov. Find the decisions most like a given one, filtered by outcome, agency, year, contracting vehicle (IDIQ, task order, set-aside and others) or record type. Or read a stored comparison of a decision with its five closest matches: what they share, how they differ, and grounds to distinguish them. Built for contracting officers and specialists, small business advocates, protest counsel, and anyone preparing or answering a bid protest. Good to know: - Summaries, issues, key points and comparisons are model-generated from the decisions. The outcome is read from the decision text. Cite B-numbers and check key facts in the decision itself before relying on them. - Decisions before 1990 come from scanned text and can have errors in names and dates. - The tools only read. There is no account, and nothing you ask changes anything. - This is a research aid, not legal advice. AcqAgent is an independent, open-source project. It is not affiliated with or endorsed by the U.S. Government Accountability Office (GAO) or any other federal agency.

Bidaya helps architecture, engineering, and construction teams find firm knowledge, discover RFPs and tenders, and manage pursuits. Search past proposals, firm documents, resumes, project experience, and company records. Find relevant opportunities and organization recommendations. Review pursuit summaries and details, and update pursuit stages and outcomes when requested. Send tool feedback to the Bidaya team when requested. Access to private records is limited to the connected user's organizations.

Access your BidPrime leads inboxes including bids, meeting keyword alerts and emerging opportunities - including matching document content for summarizing and evaluating new bids and opportunities.

BidSkim gives your AI assistant direct access to UK public procurement intelligence — the same engine behind bidskim.com. Ask which contracts in your space expire in the next six months and who holds them, find live tenders that fit what you do, and pull buyer or supplier histories on demand. Eight read-only tools: • Search live tenders by keyword, CPV code, region, value and deadline — every UK source (Find a Tender, Contracts Finder, PCS, Sell2Wales), deduplicated to one record per opportunity • Renewal intelligence: awarded contracts likely to re-tender within 24 months, with expiry dates, incumbent suppliers, recurrence evidence, required certifications and TUPE flags • Buyer profiles: spend history, top suppliers, upcoming renewals • Supplier profiles: contracts won, buyers served, current incumbencies • Meaning-based search that finds tenders which never use your keywords • Your own pipeline: AI-scored matches with reasoning, saved tenders, tracked suppliers Everything is read-only and scoped to your BidSkim account. Connections use OAuth against your existing BidSkim login — no separate credentials. Sensible daily budgets keep responses fast; tool arguments are never stored. Requires a BidSkim account on the Predict plan (bidskim.com/pricing). Setup guide: bidskim.com/mcp