Chatbot building MCP servers

A server here builds, trains or deploys AI chatbots or assistants for a website or business.

Not here: using a chatbot.

255servers
172 remote83 localTagged by Odel's AI, which can make mistakes

Best fit

98 made for this, ranked by starsRemoteLocal
  1. FlowCastle @flowcastle/flowcastleBuild, edit, and deploy Telegram bots on FlowCastle's hosted visual flow platform.4% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.2 GitHub stars.
  2. Salesforce Metadata @semwalajay83-sem/salesforce-metadata-mcpThe only Salesforce MCP with Agentforce, OmniStudio & DevOps Center tools — 228 total.7% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.2 GitHub stars.
  3. Pickaxe @aplaceforallmystuff/mcp-pickaxeMCP server for Pickaxe API - manage AI agents, knowledge bases, users, and analytics5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.2 GitHub stars.
  4. AgentStack (agentstack.tech) @agentstacktech/agentstackBackend for AI agents: one batched tool over ~668 actions (hosting, storage, auth, payments).5% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.1 GitHub stars.
  5. Zelinqa MCP @zelinqa/zelinqa-mcpGoal-oriented conversations and next-question selection through the Zelinqa API.5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.1 GitHub stars.
  6. Banking Assistant (Metabind demo) @metabind/banking-assistantDemo personal-finance MCP App: spending, subscriptions, and net worth as interactive UI.5% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; it needs your API key.1 GitHub stars.
  7. Dingdawg Support Agent @dingdawg/dingdawg-support-agentAI support agent — ticket classification, response drafting, escalation routing.6% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.1 GitHub stars.
  8. Sublime @sublime/sublimeCrie, avalie, calibre e publique agentes procedurais confiáveis em produção na Sublime.9% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; you sign in with your account.1 GitHub stars.
  9. Ar Agents MCP @ar-agents/mcpArgentine business automation: Mercado Pago, AFIP/ARCA, WhatsApp, banking, shipping (7 packages).5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.1 GitHub stars.
  10. ProAgentStore @proagentstore/platformMCP-first control plane for ProAgentStore agents and private instances.9% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; you sign in with your account.1 GitHub stars.
  11. Ateam @ariekogan/ateam-mcpBuild, validate, and deploy multi-agent AI solutions from any AI environment.11% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; you sign in with your account.1 GitHub stars.
  12. Famulor @famulor/famulor-mcpManage AI assistants, history, calls, campaigns, contacts, knowledge, messaging, and automations.6% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; you sign in with your account.1 GitHub stars.
  13. CallChatSyn @callchatsyn/callchatsynAnswer small-business customer questions from FAQs, list open appointment times, and book them.4% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
  14. Falcon Builder @falconbuilder/falcon-builderBuild, debug, and publish production AI agent workflows in Falcon Builder from any MCP client.7% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; you sign in with your account.
  15. Dialogbrain @dialogbrain/dialogbrainYour customer inbox as tools: read and answer WhatsApp, Telegram, Instagram and e-mail.10% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; you sign in with your account.
  16. Canvas @avatargaia/canvas-mcpTeamAgent Canvas MCP(服务端 0.4.0):91 个工具——开岗、改配置、跑动作、看数据、发布并配知识库。可匿名试用,带 Key 解锁全部。8% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
  17. Copilot Studio MCP @jgt87/copilot-studio-mcpBuild, test, ship and maintain Microsoft Copilot Studio agents from the editor5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.
  18. HeySale — AI Website Salesperson @heysale/heysaleCreate and install an AI Website Salesperson for sales, qualification, and lead capture.7% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; no sign-in.
  19. SiteGPT @sitegpt/sitegptManage SiteGPT chatbots and account resources through the SiteGPT API v2.9% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; you sign in with your account.
  20. PERSONAIZER @personaizer/mcpFree MCP API to create AI chat personas for websites — upload knowledge, search, and chat9% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; it needs your API key.
  21. Grounded Support Agent @ankit512/grounded-support-agentGoverned support MCP: answers only KB-grounded questions with citations, escalates the rest.3% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.
  22. Inflectiv @inflectiv/inflectiv-mcp-serverSearch, build and manage Inflectiv datasets and AI agents, plus ChainAware on-chain intelligence.5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.
  23. AskMarcel HVAC Technical Knowledge @askmarcel/mcphvac15 HVAC MCP tools: lookups, SKU provenance, guided PAC, Beta Harness sessions. 126 brands.5% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; you sign in with your account.
  24. BranderUX MCP @branderux/mcpOne message in, a full agentic application out: website and MCP app, live. Built from any AI client.10% chance this is malicious. Assessed by AI, which can make mistakes.A remote server; you sign in with your account.