[{"data":1,"prerenderedAt":1631},["ShallowReactive",2],{"mcp-directory":3},[4,442,837,1240],{"id":5,"title":6,"category":7,"decisionMatrix":8,"extension":59,"faqs":60,"items":73,"meta":424,"metaDescription":425,"metaTitle":426,"ogImage":427,"principles":428,"slug":438,"stem":439,"updatedAt":440,"__hash__":441},"mcp/mcp/customer-support.yml","Top 10 Customer Support & CRM MCP Servers for AI Agents (2026)","Customer Support & CRM",[9,14,19,24,29,34,39,44,49,54],{"step":10,"trigger":11,"recommendation":12,"reason":13},"01","You need unified cross-functional issue triage linking customer tickets to Jira bug backlogs and Confluence docs","mcp-atlassian","Provides the deepest integration across Jira Service Management and engineering issue queues with strict read-only controls.",{"step":15,"trigger":16,"recommendation":17,"reason":18},"02","You require enterprise-grade write safety, dry-run simulation, and explicit confirmation before sending customer replies","help-scout-mcp","Enforces dual confirmation flags, operation description schemas, and preview runs on all external communications.",{"step":20,"trigger":21,"recommendation":22,"reason":23},"03","You need to query enterprise Salesforce Service Cloud records, Cases, and custom CRM objects via SOQL","salesforce-mcp","The official vendor-maintained server backed by local CLI authentication and rate-limited token buckets.",{"step":25,"trigger":26,"recommendation":27,"reason":28},"04","You operate a Freshdesk helpdesk and need to manage tickets, agent notes, canned responses, and solution articles","freshdesk-mcp","Delivers comprehensive 59-tool coverage across all core Freshdesk support, contact, and knowledge base resources.",{"step":30,"trigger":31,"recommendation":32,"reason":33},"05","You manage sales deal pipelines and need optimistic concurrency protection to prevent accidental deal stage overwrites","pipedrive-mcp","Combines Bottleneck request throttling with optimistic stage concurrency checks and preview-first mutations.",{"step":35,"trigger":36,"recommendation":37,"reason":38},"06","You run a self-hosted or open-source CRM stack and need contact, company, and opportunity management","twenty-crm-mcp","Offers dedicated REST grammar support and injection-safe queries for the Twenty CRM platform.",{"step":40,"trigger":41,"recommendation":42,"reason":43},"07","You handle enterprise IT service management, incident resolutions, and change request approvals","servicenow-mcp","Supplies 82 enterprise tools spanning ServiceNow incidents, catalog items, and change request workflows.",{"step":45,"trigger":46,"recommendation":47,"reason":48},"08","You want lightweight Zendesk ticket triage and response drafting with per-operator OAuth security","zendesk-mcp","Maintains a compact 7-tool surface with knowledge base caching and OAuth token file support.",{"step":50,"trigger":51,"recommendation":52,"reason":53},"09","You need to triage real-time customer live chat inboxes sorted by longest visitor wait time","crisp-mcp","Includes specialized triage queries and rich context dossiers specifically designed for live chat operators.",{"step":55,"trigger":56,"recommendation":57,"reason":58},"10","You want semantic vector search across customer records without exceeding third-party API rate quotas","hubspot-mcp","Embeds a local FAISS vector index to search cached conversation threads and contact profiles.","yml",[61,64,67,70],{"question":62,"answer":63},"How do support MCP servers handle customer reply safety and prevent accidental email sends?","Servers vary significantly in their write safety architecture. High-safety implementations like `help-scout-mcp` disable customer-visible writes by default, require explicit runtime environment flags, and demand per-call confirmation tokens. In contrast, servers such as `freshdesk-mcp` expose public reply methods immediately upon authentication, making host-level permission controls and human approval gates essential before deploying agents with write access.",{"question":65,"answer":66},"Can MCP servers access customer session history without consuming all model context window tokens?","Long customer ticket threads can easily exhaust context windows. Servers like `crisp-mcp` provide composite tools like `get_rich_context` to consolidate essential metadata, while `mcp-atlassian` allows pagination and comment limits via JQL parameters. For large conversation histories, agents should query recent summaries before retrieving full discussion trees.",{"question":68,"answer":69},"What authentication methods do CRM and helpdesk MCP servers require?","Most servers accept API keys or personal access tokens passed through environment variables. However, servers like `salesforce-mcp` leverage existing local CLI session authentications, while `zendesk-mcp` supports per-operator OAuth 2.0 PKCE token files to preserve individual agent audit trails.",{"question":71,"answer":72},"How does Webfuse complement customer support and CRM MCP servers?","While support MCP servers provide structured API access to ticket records and CRM databases, Webfuse provides real-time web session virtualization and zero-code co-browsing. Support agents can reproduce user-reported errors, inspect live client session DOM snapshots, and guide customers through complex web workflows with human-in-the-loop checkpoints.",[74,113,149,184,217,252,287,322,355,389],{"rank":10,"slug":12,"name":12,"author":75,"repo":76,"githubUrl":77,"stars":78,"transport":79,"ogImage":80,"installCommand":81,"bestFor":82,"boundary":83,"boundaryNote":84,"summary":85,"tools":86,"configSnippet":103,"limitations":104,"webfuseSynergy":107,"runtimes":108},"sooperset","sooperset/mcp-atlassian","https://github.com/sooperset/mcp-atlassian","5.8k","stdio / Streamable HTTP","https://opengraph.githubassets.com/1/sooperset/mcp-atlassian","uvx mcp-atlassian","Cross-functional customer issue triage across Jira, Jira Service Management, and Confluence documentation","Service Desk & Issue Tracking","Bridges customer-facing support requests directly into engineering issue tracking and knowledge base spaces.","The most widely deployed Atlassian Model Context Protocol server. Built on FastMCP with 25 modular toolsets, it connects support agents to Jira Service Management queues, issue transitions, and Confluence articles with fine-grained read-only flags.",[87,91,95,99],{"name":88,"schema":89,"description":90},"jira_search","{ \"jql\": \"string\", \"limit\": \"number?\", \"fields\": \"string?\" }","Search issues using JQL syntax with field filtering.",{"name":92,"schema":93,"description":94},"jira_get_issue","{ \"issue_key\": \"string\", \"expand\": \"string?\", \"comment_limit\": \"number?\" }","Retrieve issue details, SLA status, and recent comments.",{"name":96,"schema":97,"description":98},"jira_create_customer_request","{ \"service_desk_id\": \"string\", \"request_type_id\": \"string\", \"request_field_values\": \"object\" }","Submit a new customer service desk request.",{"name":100,"schema":101,"description":102},"jira_add_comment","{ \"issue_key\": \"string\", \"comment\": \"string\", \"is_internal\": \"boolean?\" }","Add a public reply or internal work note to a ticket.","{\n  \"mcpServers\": {\n    \"atlassian\": {\n      \"command\": \"uvx\",\n      \"args\": [\"mcp-atlassian\"],\n      \"env\": {\n        \"JIRA_URL\": \"https://your-domain.atlassian.net\",\n        \"JIRA_USERNAME\": \"agent@example.com\",\n        \"JIRA_API_TOKEN\": \"your-jira-api-token\",\n        \"READ_ONLY_MODE\": \"false\"\n      }\n    }\n  }\n}\n",[105,106],"Exposing all 98 tools without filtering toolsets consumes excessive tool definition tokens.","Attachments and issue images return base64 payloads that increase token usage.","Connect Jira support tickets to Webfuse live browser sessions so support agents can inspect the exact customer browser state.",[109,110,111,112],"Claude Desktop","Cursor","OpenCode","VS Code Copilot",{"rank":15,"slug":17,"name":114,"author":115,"repo":116,"githubUrl":117,"stars":118,"transport":119,"ogImage":120,"installCommand":121,"bestFor":122,"boundary":123,"boundaryNote":124,"summary":125,"tools":126,"configSnippet":143,"limitations":144,"webfuseSynergy":147,"runtimes":148},"help-scout-mcp-server","Drew Burchfield","drewburchfield/help-scout-mcp-server","https://github.com/drewburchfield/help-scout-mcp-server","47","stdio","https://opengraph.githubassets.com/1/drewburchfield/help-scout-mcp-server","npx -y help-scout-mcp-server","High-safety helpdesk operations with explicit write gating, confirmation requirements, and dry-run modes","Helpdesk & Knowledge Base","Wraps customer conversation threads, mailboxes, and Docs sites behind a token-efficient gateway architecture.","A TypeScript MCP server designed for Help Scout mailboxes. It uses a 3-tool dynamic gateway over 55 read operations and enforces strict confirmation tokens and dry-run simulations on all outbound replies.",[127,131,135,139],{"name":128,"schema":129,"description":130},"search_help_scout","{ \"query\": \"string\" }","Find available Help Scout operations matching support intents.",{"name":132,"schema":133,"description":134},"describe_help_scout","{ \"names\": \"array\" }","Retrieve exact JSON input schemas for selected operations.",{"name":136,"schema":137,"description":138},"read_help_scout","{ \"name\": \"string\", \"arguments\": \"object\" }","Execute read queries across conversations, customers, tags, or reports.",{"name":140,"schema":141,"description":142},"write_help_scout","{ \"name\": \"string\", \"arguments\": \"object\", \"confirmedForConversationId\": \"number?\", \"dryRun\": \"boolean?\" }","Execute mutations with conversation ID matching and optional dry-run preview.","{\n  \"mcpServers\": {\n    \"helpscout\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"help-scout-mcp-server\"],\n      \"env\": {\n        \"HELPSCOUT_APP_ID\": \"your-app-id\",\n        \"HELPSCOUT_APP_SECRET\": \"your-app-secret\",\n        \"HELPSCOUT_ENABLE_WRITES\": \"true\",\n        \"HELPSCOUT_ENABLE_CUSTOMER_VISIBLE_WRITES\": \"false\"\n      }\n    }\n  }\n}\n",[145,146],"Gateway pattern requires multiple tool call turns to search, describe, and execute unfamiliar operations.","Requires OAuth client credentials and acts as a shared application identity across all sessions.","Combine Help Scout ticket triage with Webfuse session recordings to verify customer issues before sending replies.",[109,110,111],{"rank":20,"slug":22,"name":150,"author":151,"repo":152,"githubUrl":153,"stars":154,"transport":119,"ogImage":155,"installCommand":156,"bestFor":157,"boundary":158,"boundaryNote":159,"summary":160,"tools":161,"configSnippet":178,"limitations":179,"webfuseSynergy":182,"runtimes":183},"@salesforce/mcp","Salesforce (Official)","salesforcecli/mcp","https://github.com/salesforcecli/mcp","464","https://opengraph.githubassets.com/1/salesforcecli/mcp","npx -y @salesforce/mcp --orgs DEFAULT_TARGET_ORG --toolsets all","Enterprise Salesforce org administration, SOQL querying, and Service Cloud metadata operations","Enterprise CRM Platform","Provides official CLI-authenticated access to Salesforce schemas, objects, SOQL queries, and permissions.","The official Salesforce Model Context Protocol server. Running on Node with built-in token-bucket rate limiting, it uses local Salesforce CLI authentications to query records, describe objects, and manage configurations.",[162,166,170,174],{"name":163,"schema":164,"description":165},"run_soql_query","{ \"query\": \"string\", \"targetOrg\": \"string?\" }","Execute SOQL queries across Cases, Contacts, Accounts, and custom objects.",{"name":167,"schema":168,"description":169},"list_all_orgs","{ \"all\": \"boolean?\" }","List authenticated Salesforce scratch orgs, sandboxes, and production environments.",{"name":171,"schema":172,"description":173},"list_tools","{}","Query dynamically registered tools and active toolsets.",{"name":175,"schema":176,"description":177},"enable_tools","{ \"tools\": \"array\" }","Dynamically enable specific Salesforce tooling operations on demand.","{\n  \"mcpServers\": {\n    \"salesforce\": {\n      \"command\": \"npx\",\n      \"args\": [\n        \"-y\",\n        \"@salesforce/mcp\",\n        \"--orgs\",\n        \"DEFAULT_TARGET_ORG\",\n        \"--toolsets\",\n        \"data,users,metadata\"\n      ]\n    }\n  }\n}\n",[180,181],"Focuses primarily on developer tooling and SOQL rather than high-level CRM case workflow helpers.","Requires pre-authenticated org credentials managed via the Salesforce CLI.","Trigger Webfuse web automation tasks directly from customer records identified through Salesforce SOQL lookups.",[109,110,111],{"rank":25,"slug":27,"name":27,"author":185,"repo":186,"githubUrl":187,"stars":188,"transport":119,"ogImage":189,"installCommand":190,"bestFor":191,"boundary":192,"boundaryNote":193,"summary":194,"tools":195,"configSnippet":211,"limitations":212,"webfuseSynergy":215,"runtimes":216},"Gopi Krishnan & Maanaesh Swamy","effytech/freshdesk_mcp","https://github.com/effytech/freshdesk_mcp","68","https://opengraph.githubassets.com/1/effytech/freshdesk_mcp","uvx freshdesk-mcp","Comprehensive Freshdesk ticket management, customer lookups, canned responses, and solution articles","Multichannel Customer Support","Full API v2 coverage across tickets, conversations, agent groups, and solution knowledge bases.","A Python FastMCP server delivering 59 endpoints for Freshdesk. It allows agents to search tickets, inspect conversation threads, draft replies, add internal notes, and query company directories.",[196,199,203,207],{"name":197,"schema":129,"description":198},"search_tickets","Search support tickets by query filters, email, or status.",{"name":200,"schema":201,"description":202},"get_ticket_conversation","{ \"ticket_id\": \"number\" }","Fetch all incoming messages, outgoing replies, and internal notes on a ticket.",{"name":204,"schema":205,"description":206},"create_ticket_reply","{ \"ticket_id\": \"number\", \"body\": \"string\" }","Post a customer-visible reply to an existing support conversation.",{"name":208,"schema":209,"description":210},"create_ticket_note","{ \"ticket_id\": \"number\", \"body\": \"string\", \"private\": \"boolean?\" }","Attach an internal collaboration note visible only to agents.","{\n  \"mcpServers\": {\n    \"freshdesk\": {\n      \"command\": \"uvx\",\n      \"args\": [\"freshdesk-mcp\"],\n      \"env\": {\n        \"FRESHDESK_API_KEY\": \"your-freshdesk-api-key\",\n        \"FRESHDESK_DOMAIN\": \"your-company.freshdesk.com\"\n      }\n    }\n  }\n}\n",[213,214],"Does not implement proactive rate limit throttling on heavily throttled search endpoints.","Returns full un-truncated conversation bodies that consume context on long email chains.","Equip Freshdesk support agents with Webfuse browser replays to recreate client bugs directly from ticket notes.",[109,110,111],{"rank":30,"slug":32,"name":218,"author":219,"repo":220,"githubUrl":221,"stars":222,"transport":79,"ogImage":223,"installCommand":224,"bestFor":225,"boundary":226,"boundaryNote":227,"summary":228,"tools":229,"configSnippet":246,"limitations":247,"webfuseSynergy":250,"runtimes":251},"pipedrive-mcp-server","Will Dent","WillDent/pipedrive-mcp-server","https://github.com/WillDent/pipedrive-mcp-server","60","https://opengraph.githubassets.com/1/WillDent/pipedrive-mcp-server","npm install -g pipedrive-mcp-server","Sales pipeline tracking and CRM deal management with concurrency checks and proactive rate limiting","Sales Pipeline & Deals","Wraps Pipedrive deals, organizations, and persons with optimistic concurrency and Bottleneck throttling.","A Node MCP server built with Bottleneck throttling, dual stdio/HTTP transport, and preview-first writes. It lets agents search deals, check pipelines, inspect customer notes, and update deal stages with optimistic concurrency protection.",[230,234,238,242],{"name":231,"schema":232,"description":233},"get-deals","{ \"status\": \"string?\", \"stageId\": \"number?\", \"limit\": \"number?\", \"cursor\": \"string?\" }","Retrieve deals with stage filters, value ranges, and cursor pagination.",{"name":235,"schema":236,"description":237},"get-deal-notes","{ \"dealId\": \"number\", \"limit\": \"number?\" }","Fetch customer notes and communication logs for a specific deal.",{"name":239,"schema":240,"description":241},"search-all","{ \"term\": \"string\", \"itemTypes\": \"array?\" }","Search across deals, persons, organizations, leads, and files.",{"name":243,"schema":244,"description":245},"move-deal","{ \"dealId\": \"number\", \"targetStageId\": \"number\", \"expectedCurrentStageId\": \"number?\", \"execute\": \"boolean?\" }","Move a deal to a new stage with optimistic concurrency checks and preview mode.","{\n  \"mcpServers\": {\n    \"pipedrive\": {\n      \"command\": \"pipedrive-mcp-server\",\n      \"args\": [],\n      \"env\": {\n        \"PIPEDRIVE_API_TOKEN\": \"your-api-token\",\n        \"PIPEDRIVE_WRITE_TOOLS\": \"move-deal,add-deal-note\"\n      }\n    }\n  }\n}\n",[248,249],"Read tools are intentionally scoped to core pipeline entities and exclude custom deal metadata creation.","Requires global npm installation or Docker execution as it does not publish an npx binary.","Enrich Pipedrive deal records with customer onboarding milestones tracked inside Webfuse co-browsing sessions.",[109,110,111],{"rank":35,"slug":37,"name":253,"author":254,"repo":255,"githubUrl":256,"stars":257,"transport":119,"ogImage":258,"installCommand":259,"bestFor":260,"boundary":261,"boundaryNote":262,"summary":263,"tools":264,"configSnippet":281,"limitations":282,"webfuseSynergy":285,"runtimes":286},"twenty-crm-mcp-server","mhenry3164","mhenry3164/twenty-crm-mcp-server","https://github.com/mhenry3164/twenty-crm-mcp-server","100","https://opengraph.githubassets.com/1/mhenry3164/twenty-crm-mcp-server","npx -y github:mhenry3164/twenty-crm-mcp-server","Open-source CRM data operations, customer contact management, and self-hosted CRM automation","Open-Source CRM","Integrates with Twenty CRM REST APIs for people, companies, opportunities, and task tracking.","An MCP server for the Twenty open-source CRM platform. It features native REST filtering grammar, injection-sanitized search queries, cursor pagination, and backoff retries across contacts, companies, and opportunity records.",[265,269,273,277],{"name":266,"schema":267,"description":268},"list_people","{ \"search\": \"string?\", \"limit\": \"number?\", \"filter\": \"object?\", \"companyId\": \"string?\" }","List and filter customer contacts with relational depth controls.",{"name":270,"schema":271,"description":272},"create_person","{ \"firstName\": \"string\", \"lastName\": \"string\", \"email\": \"string?\", \"companyId\": \"string?\" }","Create a new person record with associated company links.",{"name":274,"schema":275,"description":276},"list_companies","{ \"search\": \"string?\", \"limit\": \"number?\", \"filter\": \"object?\" }","Query company records, revenue metrics, and ideal customer profile data.",{"name":278,"schema":279,"description":280},"create_opportunity","{ \"name\": \"string\", \"stage\": \"string\", \"amount\": \"number?\", \"companyId\": \"string?\" }","Create a new sales opportunity attached to an account.","{\n  \"mcpServers\": {\n    \"twenty\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"github:mhenry3164/twenty-crm-mcp-server\"],\n      \"env\": {\n        \"TWENTY_API_KEY\": \"your-twenty-api-key\",\n        \"TWENTY_BASE_URL\": \"https://api.twenty.com\"\n      }\n    }\n  }\n}\n",[283,284],"Workspace API keys grant administrative read and write access without granular per-user role restrictions.","Batch creation operations are capped at 60 records and are not executed in atomic transactions.","Synchronize user engagement data from Webfuse interactive web spaces into Twenty CRM customer profiles.",[109,110,111],{"rank":40,"slug":42,"name":42,"author":288,"repo":289,"githubUrl":290,"stars":291,"transport":292,"ogImage":293,"installCommand":294,"bestFor":295,"boundary":296,"boundaryNote":297,"summary":298,"tools":299,"configSnippet":316,"limitations":317,"webfuseSynergy":320,"runtimes":321},"echelon-ai-labs","echelon-ai-labs/servicenow-mcp","https://github.com/echelon-ai-labs/servicenow-mcp","297","stdio / SSE","https://opengraph.githubassets.com/1/echelon-ai-labs/servicenow-mcp","pip install -e .","Enterprise IT service management (ITSM), incident resolution, change approvals, and catalog requests","Enterprise ITSM & Incident Management","Bridges enterprise IT support, change request lifecycles, and Table API resources into agent contexts.","An enterprise-grade MCP server for ServiceNow supporting both stdio and SSE transports. With 82 tools across Table APIs, it handles incident creation, assignment updates, work notes, change approvals, and knowledge base lookups.",[300,304,308,312],{"name":301,"schema":302,"description":303},"list_incidents","{ \"limit\": \"number?\", \"state\": \"string?\", \"assigned_to\": \"string?\", \"query\": \"string?\" }","Query active IT incidents by priority, assignment group, or state.",{"name":305,"schema":306,"description":307},"create_incident","{ \"short_description\": \"string\", \"urgency\": \"string?\", \"impact\": \"string?\", \"caller_id\": \"string?\" }","Create a new incident record in ServiceNow.",{"name":309,"schema":310,"description":311},"add_comment","{ \"sys_id\": \"string\", \"comment\": \"string\", \"is_work_note\": \"boolean?\" }","Append a public comment or internal work note to an incident.",{"name":313,"schema":314,"description":315},"approve_change","{ \"change_sys_id\": \"string\", \"comment\": \"string?\" }","Submit an approval vote on an active change request.","{\n  \"mcpServers\": {\n    \"servicenow\": {\n      \"command\": \"servicenow-mcp\",\n      \"args\": [],\n      \"env\": {\n        \"SERVICENOW_INSTANCE_URL\": \"https://your-instance.service-now.com\",\n        \"SERVICENOW_AUTH_TYPE\": \"basic\",\n        \"SERVICENOW_USERNAME\": \"admin\",\n        \"SERVICENOW_PASSWORD\": \"your-password\"\n      }\n    }\n  }\n}\n",[318,319],"Workflow script tools require custom Script Execution API endpoint configuration on production instances.","Pydantic string error formatting loses structured diagnostic context during Table API validation errors.","Resolve complex enterprise support incidents by opening Webfuse virtual sessions directly from ServiceNow records.",[109,110,111],{"rank":45,"slug":47,"name":323,"author":324,"repo":325,"githubUrl":326,"stars":327,"transport":119,"ogImage":328,"installCommand":329,"bestFor":330,"boundary":331,"boundaryNote":332,"summary":333,"tools":334,"configSnippet":349,"limitations":350,"webfuseSynergy":353,"runtimes":354},"zendesk-mcp-server","reminia","reminia/zendesk-mcp-server","https://github.com/reminia/zendesk-mcp-server","115","https://opengraph.githubassets.com/1/reminia/zendesk-mcp-server","uvx zendesk-mcp-server","Lightweight Zendesk ticket inspection, customer reply drafting, and help center knowledge base lookups","Customer Helpdesk & Ticketing","Simple, predictable stdio bridge to Zendesk tickets, comments, and knowledge base resources.","A focused Python MCP server for Zendesk Support. It provides tools to retrieve ticket details, inspect comment threads, draft responses, and read cached knowledge base articles with per-operator OAuth audit trails.",[335,339,342,345],{"name":336,"schema":337,"description":338},"get_tickets","{ \"page\": \"number?\", \"per_page\": \"number?\", \"sort_by\": \"string?\" }","List recent tickets with pagination parameters.",{"name":340,"schema":201,"description":341},"get_ticket","Retrieve full details and custom fields for a specific ticket.",{"name":343,"schema":201,"description":344},"get_ticket_comments","Fetch the chronological discussion history on a support ticket.",{"name":346,"schema":347,"description":348},"create_ticket_comment","{ \"ticket_id\": \"number\", \"comment\": \"string\", \"public\": \"boolean?\" }","Add a public customer reply or internal note to a Zendesk ticket.","{\n  \"mcpServers\": {\n    \"zendesk\": {\n      \"command\": \"uvx\",\n      \"args\": [\"zendesk-mcp-server\"],\n      \"env\": {\n        \"ZENDESK_SUBDOMAIN\": \"your-subdomain\",\n        \"ZENDESK_CLIENT_ID\": \"your-oauth-client-id\"\n      }\n    }\n  }\n}\n",[351,352],"Lacks a dedicated search endpoint, requiring sequential pagination over ticket lists.","Does not include built-in automatic backoff when encountering Zendesk API rate limits.","Provide Zendesk support agents with verified DOM snapshots of customer sessions captured through Webfuse.",[109,110,111],{"rank":50,"slug":52,"name":52,"author":356,"repo":357,"githubUrl":358,"stars":359,"transport":119,"ogImage":360,"installCommand":361,"bestFor":362,"boundary":363,"boundaryNote":364,"summary":365,"tools":366,"configSnippet":383,"limitations":384,"webfuseSynergy":387,"runtimes":388},"zernio-dev","zernio-dev/crisp-mcp","https://github.com/zernio-dev/crisp-mcp","13","https://opengraph.githubassets.com/1/zernio-dev/crisp-mcp","node dist/index.js","Real-time live chat triage, inbox message routing, and operator assignment across Crisp websites","Live Chat & Real-Time Inbox","Connects agents to live visitor sessions, inbox queues, operator routing, and composite context dossiers.","A Node MCP server tailored for live chat customer support in Crisp. It exposes 33 tools and 2 MCP resources, enabling agents to query unread threads, assemble customer context dossiers, and dispatch live responses.",[367,371,375,379],{"name":368,"schema":369,"description":370},"conversations_awaiting_reply","{ \"max_pages\": \"number?\" }","Retrieve live chat conversations prioritized by longest customer wait time.",{"name":372,"schema":373,"description":374},"get_rich_context","{ \"session_id\": \"string\" }","Fetch a complete customer dossier combining messages, profile details, and session metadata.",{"name":376,"schema":377,"description":378},"send_message","{ \"session_id\": \"string\", \"content\": \"string\", \"type\": \"string?\" }","Send a live chat response or internal note to a visitor.",{"name":380,"schema":381,"description":382},"assign_conversation","{ \"session_id\": \"string\", \"operator_id\": \"string\" }","Route a live conversation to a specific support operator.","{\n  \"mcpServers\": {\n    \"crisp\": {\n      \"command\": \"node\",\n      \"args\": [\"/path/to/crisp-mcp/dist/index.js\"],\n      \"env\": {\n        \"CRISP_IDENTIFIER\": \"your-plugin-identifier\",\n        \"CRISP_KEY\": \"your-plugin-key\",\n        \"CRISP_WEBSITE_ID\": \"your-website-id\"\n      }\n    }\n  }\n}\n",[385,386],"Requires building from source as it is not packaged on npm or PyPI.","Destructive actions such as conversation deletion are enabled without separate confirmation gating.","Launch Webfuse zero-code co-browsing directly inside Crisp live chat sessions to guide customers through complex web forms.",[109,110,111],{"rank":55,"slug":57,"name":390,"author":391,"repo":392,"githubUrl":393,"stars":394,"transport":119,"ogImage":395,"installCommand":396,"bestFor":397,"boundary":398,"boundaryNote":399,"summary":400,"tools":401,"configSnippet":418,"limitations":419,"webfuseSynergy":422,"runtimes":423},"mcp-server-hubspot","baryhuang","baryhuang/mcp-hubspot","https://github.com/baryhuang/mcp-hubspot","128","https://opengraph.githubassets.com/1/baryhuang/mcp-hubspot","docker run -i --rm -e HUBSPOT_ACCESS_TOKEN buryhuang/mcp-hubspot:latest","Semantic CRM searching and customer company activity retrieval using local vector indexing","Inbound Marketing & CRM","Combines HubSpot contacts, companies, and tickets with local FAISS vector embeddings.","A containerized Python MCP server integrating HubSpot CRM. It pairs standard contact and ticket APIs with a local FAISS vector cache, enabling semantic search over customer interactions while mitigating API quota consumption.",[402,406,410,414],{"name":403,"schema":404,"description":405},"hubspot_search_data","{ \"query\": \"string\", \"limit\": \"number?\" }","Perform semantic vector searches across cached customer contacts and conversation threads.",{"name":407,"schema":408,"description":409},"hubspot_get_tickets","{ \"criteria\": \"object?\", \"limit\": \"number?\", \"max_retries\": \"number?\" }","Retrieve support tickets with customizable filter criteria and retry controls.",{"name":411,"schema":412,"description":413},"hubspot_get_contact","{ \"contact_id\": \"string\" }","Fetch comprehensive properties and activity history for a specific contact.",{"name":415,"schema":416,"description":417},"hubspot_create_company","{ \"name\": \"string\", \"domain\": \"string?\", \"properties\": \"object?\" }","Create a company record with automated domain duplicate prevention.","{\n  \"mcpServers\": {\n    \"hubspot\": {\n      \"command\": \"docker\",\n      \"args\": [\n        \"run\",\n        \"-i\",\n        \"--rm\",\n        \"-e\",\n        \"HUBSPOT_ACCESS_TOKEN\",\n        \"buryhuang/mcp-hubspot:latest\"\n      ],\n      \"env\": {\n        \"HUBSPOT_ACCESS_TOKEN\": \"pat-na1-your-token\"\n      }\n    }\n  }\n}\n",[420,421],"Relies on a fixed container storage path for FAISS embedding indexes, making Docker execution necessary.","Downloads local transformer embedding models on first startup, introducing initial container launch latency.","Log customer interaction insights and browsing session summaries directly into HubSpot contact timelines.",[109,110,111],{},"The top 10 Model Context Protocol (MCP) servers for Zendesk, Jira Service Management, Salesforce, Freshdesk, HubSpot, and CRM operations, with tool schemas and configs for Claude Desktop, Cursor, and OpenCode.","Top 10 Customer Support & CRM MCP Servers for AI Agents (2026) | Webfuse","/mcp/thumbnails/customer-support.png",[429,432,435],{"title":430,"description":431},"Official protocol compliance","Evaluated for strict implementation of standard MCP transports over stdio and Streamable HTTP, supporting tool schemas and resource contracts.",{"title":433,"description":434},"Tool schemas taken from source","Every callable tool name, argument signature, launch entrypoint, and environment variable is extracted directly from the repository source code.",{"title":436,"description":437},"Context and token discipline","Assessed on how responses handle large ticket conversations, ticket histories, and nested customer objects without overflowing model context.","customer-support","mcp/customer-support","September 2026","j54C7pnNv1F-STLlKv7NLiOyqaJKf4h3yGUU2xMZs_U",{"id":443,"title":444,"category":445,"decisionMatrix":446,"extension":59,"faqs":487,"items":503,"meta":822,"metaDescription":823,"metaTitle":824,"ogImage":825,"principles":826,"slug":833,"stem":834,"updatedAt":835,"__hash__":836},"mcp/mcp/developer-tools.yml","Top 10 Developer & Coding MCP Servers for AI Agents (2026)","Developer Tools & Infrastructure",[447,451,455,459,463,467,471,475,479,483],{"step":10,"trigger":448,"recommendation":449,"reason":450},"Need repo-wide GitHub PR reviews, issue management, and CI/CD log inspection","github-mcp-server","Official GitHub tool maintaining full source-control APIs and Actions telemetry.",{"step":15,"trigger":452,"recommendation":453,"reason":454},"Agent generating hallucinated APIs from outdated documentation","upstash-context7","Pulls current, verified library docs and code patterns dynamically at query time.",{"step":20,"trigger":456,"recommendation":457,"reason":458},"Multi-file refactoring and AST symbol navigation without flooding context","oraios-serena","Tree-sitter and LSP-backed AST parser that keeps token payloads ultra-compact.",{"step":25,"trigger":460,"recommendation":461,"reason":462},"Need a controlled local shell to run builds, tests, and diff file edits","wonderwhy-desktop-commander","Leading stdio terminal executor with streaming output and safe diff application.",{"step":30,"trigger":464,"recommendation":465,"reason":466},"Querying, schema inspecting, or migrating PostgreSQL, MySQL, or Spanner databases","googleapis-mcp-toolbox","Official Google multi-database tool with connection pooling and schema introspection.",{"step":35,"trigger":468,"recommendation":469,"reason":470},"Investigating production stack traces, error spikes, and user crash breadcrumbs","getsentry-sentry-mcp","Official Sentry server connecting agents directly to live crash telemetry.",{"step":40,"trigger":472,"recommendation":473,"reason":474},"Deploying and managing edge compute, D1 databases, and DNS records","cloudflare-mcp-server","Official remote Workers-based server with comprehensive Cloudflare platform coverage.",{"step":45,"trigger":476,"recommendation":477,"reason":478},"Troubleshooting Kubernetes pods, deployments, and streaming crash-loop logs","containers-kubernetes-mcp","Red Hat-sponsored cluster manager using standard kubeconfig authentication.",{"step":50,"trigger":480,"recommendation":481,"reason":482},"Validating HCL syntax and discovering Terraform provider resource schemas","hashicorp-terraform-mcp","Official HashiCorp tool backed by the live Terraform Registry schema API.",{"step":55,"trigger":484,"recommendation":485,"reason":486},"Linking PR work, triaging engineering backlogs, and updating sprint tickets","jerhadf-linear-mcp","Full-featured Linear API wrapper for agile issue management.",[488,491,494,497,500],{"question":489,"answer":490},"How do I install and configure an MCP server in Claude Desktop or Cursor?","Copy the verified `claude_desktop_config.json` snippet from this guide into your local configuration file (`~/Library/Application Support/Claude/claude_desktop_config.json` on macOS or `%APPDATA%\\Claude\\claude_desktop_config.json` on Windows). For Cursor or OpenCode, add the server under your agent settings or `.cursorrules`.",{"question":492,"answer":493},"What is the difference between stdio and HTTP/SSE MCP transports?","`stdio` launches the MCP server as a local child process and communicates via standard input/output streams (best for local CLIs and desktop tools). `SSE` (Server-Sent Events) and remote HTTP transports communicate over HTTPS/WebSockets, allowing cloud-hosted agents to invoke tools hosted remotely without local binary dependencies.",{"question":495,"answer":496},"How do I prevent MCP servers from consuming excessive LLM context tokens?","Choose MCP servers that provide pagination or AST symbol queries (like `serena` or `desktop-commander`) rather than whole-file dumps. For database and API tools, use parameterized queries that clamp row results automatically.",{"question":498,"answer":499},"Are these MCP servers safe to run on production codebases?","All 10 are open source, so you can audit them yourself. Servers with shell access (`desktop-commander`) or write permissions (`database-toolbox`, `github`) should run with read-only credentials, or behind a human approval checkpoint that gates the destructive calls.",{"question":501,"answer":502},"Can I run multiple MCP servers simultaneously in one agent session?","Yes. MCP hosts like Claude Desktop, Cursor, and OpenCode allow you to register multiple servers under `mcpServers`. The LLM automatically routes specific queries to the matching server based on tool descriptions.",[504,538,567,598,630,662,694,726,758,788],{"rank":10,"slug":449,"name":505,"author":506,"repo":505,"githubUrl":507,"stars":508,"transport":509,"ogImage":510,"installCommand":511,"bestFor":512,"boundary":513,"boundaryNote":514,"summary":515,"tools":516,"configSnippet":532,"limitations":533,"webfuseSynergy":536,"runtimes":537},"github/github-mcp-server","GitHub (Official)","https://github.com/github/github-mcp-server","32.6k","HTTP (Remote) / stdio (Docker)","https://opengraph.githubassets.com/1/github/github-mcp-server","docker run -i --rm -e GITHUB_PERSONAL_ACCESS_TOKEN ghcr.io/github/github-mcp-server","Full GitHub repository operations, pull request automation, code search, and Actions CI/CD triaging","Source Control & CI/CD","Connects agents directly to GitHub platform APIs for PR reviews, issue management, and workflow logs.","The official GitHub-maintained Model Context Protocol server. Written in Go, it allows AI coding agents to search code across repositories, inspect commit histories, manage pull requests, create issues, and query GitHub Actions run logs.",[517,520,524,528],{"name":518,"schema":129,"description":519},"search_repositories","Search for GitHub repositories with language, star, and owner qualifiers.",{"name":521,"schema":522,"description":523},"get_file_contents","{ \"owner\": \"string\", \"repo\": \"string\", \"path\": \"string\", \"ref\": \"string?\" }","Fetch file or directory contents from a specified branch or commit.",{"name":525,"schema":526,"description":527},"create_pull_request","{ \"owner\": \"string\", \"repo\": \"string\", \"title\": \"string\", \"head\": \"string\", \"base\": \"string\", \"body\": \"string?\" }","Create a new pull request across branches or forks.",{"name":529,"schema":530,"description":531},"list_workflow_runs","{ \"owner\": \"string\", \"repo\": \"string\", \"workflow_id\": \"string?\" }","Retrieve recent GitHub Actions CI/CD workflow run statuses and failure logs.","{\n  \"mcpServers\": {\n    \"github\": {\n      \"command\": \"docker\",\n      \"args\": [\n        \"run\",\n        \"-i\",\n        \"--rm\",\n        \"-e\",\n        \"GITHUB_PERSONAL_ACCESS_TOKEN\",\n        \"ghcr.io/github/github-mcp-server\"\n      ],\n      \"env\": {\n        \"GITHUB_PERSONAL_ACCESS_TOKEN\": \"ghp_yourTokenHere\"\n      }\n    }\n  }\n}\n",[534,535],"Requires a GitHub Personal Access Token (PAT) with broad repository permissions.","Large repo-wide code searches can return extensive JSON responses that consume significant context tokens.","Put a human approval checkpoint in the Webfuse session so an engineer signs off before an automated PR merge or issue closure runs.",[109,110,111,112],{"rank":15,"slug":453,"name":539,"author":540,"repo":541,"githubUrl":542,"stars":543,"transport":544,"ogImage":545,"installCommand":546,"bestFor":547,"boundary":548,"boundaryNote":549,"summary":550,"tools":551,"configSnippet":560,"limitations":561,"webfuseSynergy":564,"runtimes":565},"@upstash/context7-mcp","Upstash","upstash/context7","https://github.com/upstash/context7","61.4k","stdio / HTTP Remote","https://opengraph.githubassets.com/1/upstash/context7","npx -y @upstash/context7-mcp","Injecting version-accurate, real-time library documentation and live code examples to eliminate API hallucinations","Documentation & Knowledge Grounding","Fetches structured llms.txt, API references, and verified code patterns at agent query time.","Context7 is the leading documentation retrieval MCP server. Built on TypeScript and vector search over curated developer documentation, it resolves library versions on demand so agents generate code against actual current SDK APIs rather than outdated training data.",[552,556],{"name":553,"schema":554,"description":555},"resolve_library_id","{ \"libraryName\": \"string\" }","Resolves a library package name into the canonical documentation index ID.",{"name":557,"schema":558,"description":559},"query_docs","{ \"libraryId\": \"string\", \"query\": \"string\" }","Searches official documentation and extracts concise, relevant API snippets.","{\n  \"mcpServers\": {\n    \"context7\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"@upstash/context7-mcp\"]\n    }\n  }\n}\n",[562,563],"Requires outbound network connectivity to Upstash's documentation indexing service.","Internal proprietary enterprise libraries must be indexed separately.","Grounds agents in current documentation before they invoke live Webfuse web proxy and browser control tools.",[109,110,111,566],"Codex",{"rank":20,"slug":457,"name":568,"author":569,"repo":568,"githubUrl":570,"stars":571,"transport":572,"ogImage":573,"installCommand":574,"bestFor":575,"boundary":576,"boundaryNote":577,"summary":578,"tools":579,"configSnippet":592,"limitations":593,"webfuseSynergy":596,"runtimes":597},"oraios/serena","Oraios AI (Viacheslav Kovalevskyi)","https://github.com/oraios/serena","28.7k","stdio (Python / uvx)","https://opengraph.githubassets.com/1/oraios/serena","uvx serena","Tree-sitter AST semantic code search, symbol-level navigation, and token-efficient code refactoring","AST Code Intelligence & Refactoring","Acts as a headless Language Server Protocol (LSP) and AST indexer for multi-file codebases.","Serena gives AI coding agents full semantic IDE capabilities. By combining Tree-sitter AST parsers and Language Server Protocol bindings, it allows agents to find symbol definitions, trace type hierarchies, and execute structural replacements without loading whole files into context.",[580,584,588],{"name":581,"schema":582,"description":583},"find_symbol","{ \"name\": \"string\", \"kind\": \"string?\" }","Locate class, interface, or function declarations across the entire codebase.",{"name":585,"schema":586,"description":587},"find_references","{ \"symbol\": \"string\", \"file_path\": \"string\" }","Find all call sites and usages of a specific code symbol.",{"name":589,"schema":590,"description":591},"get_ast_tree","{ \"file_path\": \"string\", \"max_depth\": \"number?\" }","Returns a compressed syntactic hierarchy of a file to reduce token overhead.","{\n  \"mcpServers\": {\n    \"serena\": {\n      \"command\": \"uvx\",\n      \"args\": [\"serena\"]\n    }\n  }\n}\n",[594,595],"Requires Python 3.10+ and a local Tree-sitter build environment.","Large monorepos require initial indexing time before symbol queries become instantaneous.","Ensures agents understand code structure before generating automation scripts that run through Webfuse proxies.",[109,110,111],{"rank":25,"slug":461,"name":599,"author":600,"repo":601,"githubUrl":602,"stars":603,"transport":604,"ogImage":605,"installCommand":606,"bestFor":607,"boundary":608,"boundaryNote":609,"summary":610,"tools":611,"configSnippet":624,"limitations":625,"webfuseSynergy":628,"runtimes":629},"@wonderwhy-er/desktop-commander","Igor Walton (wonderwhy-er)","wonderwhy-er/DesktopCommanderMCP","https://github.com/wonderwhy-er/DesktopCommanderMCP","9.5k","stdio (Node.js / npx)","https://opengraph.githubassets.com/1/wonderwhy-er/DesktopCommanderMCP","npx -y @wonderwhy-er/desktop-commander","Local shell command execution, process management, and diff-based filesystem edits","Terminal & Local Execution","Provides a controlled terminal execution loop with timeout and output streaming capabilities.","Desktop Commander is the go-to MCP server for giving AI agents a local shell. It provides structured commands to spawn terminal sessions, run background builds/tests, stream stdout/stderr, and perform surgical diff replacements on files.",[612,616,620],{"name":613,"schema":614,"description":615},"execute_command","{ \"command\": \"string\", \"timeout_ms\": \"number?\" }","Execute a shell command in the workspace with bounded timeout and exit code capture.",{"name":617,"schema":618,"description":619},"read_file","{ \"path\": \"string\", \"offset\": \"number?\", \"limit\": \"number?\" }","Read file contents with line-based pagination to preserve context window tokens.",{"name":621,"schema":622,"description":623},"edit_file_diff","{ \"path\": \"string\", \"old_string\": \"string\", \"new_string\": \"string\" }","Perform precise, targeted string replacement on local source files.","{\n  \"mcpServers\": {\n    \"desktop-commander\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"@wonderwhy-er/desktop-commander\"]\n    }\n  }\n}\n",[626,627],"Has access to local system shell; requires careful workspace path scoping to prevent unintended changes.","Long-running commands must be managed with explicit timeouts to avoid blocking agent turns.","Acts as the local execution agent companion alongside Webfuse's remote browser session execution.",[109,110,111],{"rank":30,"slug":465,"name":631,"author":632,"repo":633,"githubUrl":634,"stars":635,"transport":636,"ogImage":637,"installCommand":638,"bestFor":639,"boundary":640,"boundaryNote":641,"summary":642,"tools":643,"configSnippet":656,"limitations":657,"webfuseSynergy":660,"runtimes":661},"google/mcp-toolbox","Google (googleapis)","googleapis/mcp-toolbox","https://github.com/googleapis/mcp-toolbox","16.3k","stdio / HTTP (Go)","https://opengraph.githubassets.com/1/googleapis/mcp-toolbox","docker run -i --rm -e DB_URL google/mcp-toolbox","Universal database queries, schema inspection, and connection pooling for PostgreSQL, MySQL, Spanner, and SQL Server","Relational Databases & SQL","Enterprise database connectivity with built-in connection pooling and schema introspection.","MCP Toolbox by Google is the modern replacement for legacy, unmaintained single-database MCP servers. Written in high-performance Go, it connects AI agents safely to PostgreSQL, MySQL, Cloud Spanner, and AlloyDB with parameterized queries and schema introspection.",[644,648,652],{"name":645,"schema":646,"description":647},"list_tables","{ \"database\": \"string?\" }","Enumerate database tables, views, and structural constraints.",{"name":649,"schema":650,"description":651},"describe_table","{ \"table_name\": \"string\" }","Inspect column names, data types, primary keys, and foreign key relationships.",{"name":653,"schema":654,"description":655},"execute_query","{ \"query\": \"string\", \"params\": \"array?\" }","Execute a parameterized SQL query with automatic row limit clamping.","{\n  \"mcpServers\": {\n    \"database-toolbox\": {\n      \"command\": \"docker\",\n      \"args\": [\"run\", \"-i\", \"--rm\", \"-e\", \"DB_URL\", \"google/mcp-toolbox\"],\n      \"env\": {\n        \"DB_URL\": \"postgresql://user:password@localhost:5432/mydb\"\n      }\n    }\n  }\n}\n",[658,659],"Direct write queries must be restricted via database user permissions to prevent accidental data drops.","Very wide tables with hundreds of columns should be queried with explicit column selectors.","Ensures database state can be validated alongside end-to-end web user actions executed in Webfuse.",[109,110,111],{"rank":35,"slug":469,"name":663,"author":664,"repo":665,"githubUrl":666,"stars":667,"transport":668,"ogImage":669,"installCommand":670,"bestFor":671,"boundary":672,"boundaryNote":673,"summary":674,"tools":675,"configSnippet":688,"limitations":689,"webfuseSynergy":692,"runtimes":693},"@sentry/mcp-server","Sentry (Official)","getsentry/sentry-mcp","https://github.com/getsentry/sentry-mcp","850","stdio (Node.js) / HTTP Remote","https://opengraph.githubassets.com/1/getsentry/sentry-mcp","npx -y @sentry/mcp-server","Real-time production error triage, stack trace analysis, and Seer AI issue investigation","Observability & Error Tracking","Bridges agents to live production crash logs, breadcrumbs, and release telemetry.","The official Sentry MCP server connects coding agents directly to production telemetry. When a bug occurs, the agent can fetch the exact stack trace, affected user breadcrumbs, recent commit associations, and release health metrics directly from Sentry.",[676,680,684],{"name":677,"schema":678,"description":679},"get_issue","{ \"issue_id\": \"string\" }","Retrieve comprehensive details, frequency, and stack trace for a specific Sentry issue.",{"name":681,"schema":682,"description":683},"list_recent_errors","{ \"project\": \"string\", \"query\": \"string?\" }","Search active unhandled exceptions and performance anomalies in a project.",{"name":685,"schema":686,"description":687},"get_event_breadcrumbs","{ \"event_id\": \"string\" }","Inspect chronological user interactions, network requests, and console logs leading to an exception.","{\n  \"mcpServers\": {\n    \"sentry\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"@sentry/mcp-server\"],\n      \"env\": {\n        \"SENTRY_AUTH_TOKEN\": \"sntrys_yourTokenHere\"\n      }\n    }\n  }\n}\n",[690,691],"Requires a Sentry User Auth Token with issue read permissions.","Stack traces for minified code require sourcemaps already uploaded to Sentry.","Agents can diagnose web errors detected by Sentry, then reproduce and fix them using Webfuse live browser replays.",[109,110,111],{"rank":40,"slug":473,"name":695,"author":696,"repo":697,"githubUrl":698,"stars":699,"transport":700,"ogImage":701,"installCommand":702,"bestFor":703,"boundary":704,"boundaryNote":705,"summary":706,"tools":707,"configSnippet":720,"limitations":721,"webfuseSynergy":724,"runtimes":725},"@cloudflare/mcp-server-cloudflare","Cloudflare (Official)","cloudflare/mcp-server-cloudflare","https://github.com/cloudflare/mcp-server-cloudflare","4.1k","HTTP Remote (OAuth) / stdio","https://opengraph.githubassets.com/1/cloudflare/mcp-server-cloudflare","https://mcp.cloudflare.com/mcp","Managing Cloudflare Workers, Pages, D1 SQL, R2 Object Storage, and DNS records","Serverless & Cloud Infrastructure","Edge compute, serverless storage, and global DNS orchestration via remote OAuth MCP.","Cloudflare's official MCP server provides direct agent access to the Cloudflare developer platform. Built natively on Cloudflare Workers, it enables agents to deploy Workers, query edge D1 databases, manage R2 buckets, inspect build logs, and configure routing rules.",[708,712,716],{"name":709,"schema":710,"description":711},"list_workers","{ \"account_id\": \"string\" }","List all deployed Cloudflare Workers scripts and their active routes.",{"name":713,"schema":714,"description":715},"query_d1_database","{ \"database_id\": \"string\", \"sql\": \"string\" }","Execute SQL queries on edge-replicated Cloudflare D1 SQLite databases.",{"name":717,"schema":718,"description":719},"get_pages_deployment","{ \"project_name\": \"string\", \"deployment_id\": \"string\" }","Inspect Cloudflare Pages build logs and deployment status.","{\n  \"mcpServers\": {\n    \"cloudflare\": {\n      \"type\": \"http\",\n      \"url\": \"https://mcp.cloudflare.com/mcp\"\n    }\n  }\n}\n",[722,723],"Remote HTTP transport requires an MCP host with OAuth support (VS Code 1.101+, Cursor).","Write operations require explicit Cloudflare API account token scoping.","Deploy edge apps on Cloudflare, then check them from other regions through the Webfuse Augmented Web Proxy.",[112,110,109,111],{"rank":45,"slug":477,"name":727,"author":728,"repo":729,"githubUrl":730,"stars":731,"transport":732,"ogImage":733,"installCommand":734,"bestFor":735,"boundary":736,"boundaryNote":737,"summary":738,"tools":739,"configSnippet":752,"limitations":753,"webfuseSynergy":756,"runtimes":757},"kubernetes-mcp-server","Containers Org / Red Hat","containers/kubernetes-mcp-server","https://github.com/containers/kubernetes-mcp-server","2.0k","stdio (Node.js) / Docker","https://opengraph.githubassets.com/1/containers/kubernetes-mcp-server","npx -y kubernetes-mcp-server","Kubernetes and OpenShift cluster inspection, pod debugging, log streaming, and deployment rollouts","Containers & Kubernetes Orchestration","Cluster lifecycle management utilizing standard kubeconfig credentials.","The Kubernetes MCP server enables AI agents to inspect and troubleshoot containerized infrastructure. Using your existing kubeconfig, agents can list pods, stream failing container logs, describe deployments, and inspect namespace events without raw kubectl guessing.",[740,744,748],{"name":741,"schema":742,"description":743},"list_pods","{ \"namespace\": \"string?\" }","List running, pending, and crash-looping pods in a target namespace.",{"name":745,"schema":746,"description":747},"get_pod_logs","{ \"pod_name\": \"string\", \"namespace\": \"string?\", \"tail_lines\": \"number?\" }","Fetch tail logs from specific pod containers to diagnose crash loops.",{"name":749,"schema":750,"description":751},"describe_resource","{ \"kind\": \"string\", \"name\": \"string\", \"namespace\": \"string?\" }","Inspect Kubernetes events, resource limits, and status conditions.","{\n  \"mcpServers\": {\n    \"kubernetes\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"kubernetes-mcp-server\"]\n    }\n  }\n}\n",[754,755],"Relies on active local `~/.kube/config` context; switching clusters changes agent scope.","Mutating operations (apply/delete) should be gated with RBAC read-only service accounts.","Pair Kubernetes deployment tracking with Webfuse browser verification tests to confirm live app health.",[109,110,111],{"rank":50,"slug":481,"name":759,"author":760,"repo":759,"githubUrl":761,"stars":762,"transport":763,"ogImage":764,"installCommand":765,"bestFor":766,"boundary":767,"boundaryNote":768,"summary":769,"tools":770,"configSnippet":782,"limitations":783,"webfuseSynergy":786,"runtimes":787},"hashicorp/terraform-mcp-server","HashiCorp (Official)","https://github.com/hashicorp/terraform-mcp-server","1.5k","stdio (Go binary / Docker)","https://opengraph.githubassets.com/1/hashicorp/terraform-mcp-server","docker run -i --rm mcp/terraform-mcp-server","Terraform provider discovery, HCL module generation, and syntax schema validation","Infrastructure as Code (IaC)","Official HashiCorp Registry search and schema extraction for AWS, GCP, Azure, and Cloudflare providers.","HashiCorp's official Terraform MCP server provides AI agents with deep knowledge of Terraform providers and modules. It queries the Terraform Registry for exact argument schemas, provider versions, and valid HCL resource attributes, eliminating invalid infrastructure code generation.",[771,774,778],{"name":772,"schema":129,"description":773},"search_providers","Search official and community Terraform providers in the registry.",{"name":775,"schema":776,"description":777},"get_resource_schema","{ \"provider\": \"string\", \"resource_type\": \"string\" }","Fetch required arguments, optional attributes, and types for any Terraform resource.",{"name":779,"schema":780,"description":781},"validate_module","{ \"module_path\": \"string\" }","Verify syntax and provider constraints for a local Terraform configuration.","{\n  \"mcpServers\": {\n    \"terraform\": {\n      \"command\": \"docker\",\n      \"args\": [\"run\", \"-i\", \"--rm\", \"mcp/terraform-mcp-server\"]\n    }\n  }\n}\n",[784,785],"Focuses on provider schema and module validation; does not execute `terraform apply` directly.","Requires Docker or Go toolchain for binary execution.","Ensures cloud infrastructure provisioning matches production security standards.",[109,110,111],{"rank":55,"slug":485,"name":789,"author":790,"repo":789,"githubUrl":791,"stars":792,"transport":793,"ogImage":794,"installCommand":795,"bestFor":796,"boundary":797,"boundaryNote":798,"summary":799,"tools":800,"configSnippet":816,"limitations":817,"webfuseSynergy":820,"runtimes":821},"jerhadf/linear-mcp-server","Jeremy Hadfield","https://github.com/jerhadf/linear-mcp-server","350","stdio (TypeScript / npx)","https://opengraph.githubassets.com/1/jerhadf/linear-mcp-server","npx -y @jerhadf/linear-mcp-server","Linear project management, backlog issue search, ticket creation, and automated sprint updates","Project Management & Issue Tracking","Bridges agents to team sprint workflows, issue status transitions, and engineering backlogs.","The Linear MCP server connects coding agents directly to Linear issue tracking. Agents can search backlog tickets, create detailed bug reports with reproducible steps, transition ticket statuses, and link PR commits back to sprint deliverables.",[801,805,809,813],{"name":802,"schema":803,"description":804},"search_issues","{ \"query\": \"string\", \"teamId\": \"string?\", \"status\": \"string?\" }","Search Linear issues by title, description, assignee, and state.",{"name":806,"schema":807,"description":808},"create_issue","{ \"title\": \"string\", \"teamId\": \"string\", \"description\": \"string?\", \"priority\": \"number?\" }","Create a new issue with markdown description, priority level, and labels.",{"name":810,"schema":811,"description":812},"update_issue","{ \"id\": \"string\", \"stateId\": \"string?\", \"assigneeId\": \"string?\" }","Update issue progress, assignee, or priority.",{"name":309,"schema":814,"description":815},"{ \"issueId\": \"string\", \"body\": \"string\" }","Post a status update or automated PR summary comment to a Linear ticket.","{\n  \"mcpServers\": {\n    \"linear\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"@jerhadf/linear-mcp-server\"],\n      \"env\": {\n        \"LINEAR_API_KEY\": \"lin_api_yourKeyHere\"\n      }\n    }\n  }\n}\n",[818,819],"Requires a Linear Personal API Key with issue write access.","Team IDs must be resolved once or cached in agent context.","Automate bug ticket filing in Linear whenever Webfuse proxy or checkpoint verifications fail.",[109,110,111],{},"The top 10 Model Context Protocol (MCP) servers for Git, code search, container execution, error tracking, cloud infrastructure, and databases, with tool schemas and configs for Claude Desktop, Cursor, and OpenCode.","Top 10 Developer & Coding MCP Servers for AI Agents (2026) | Webfuse","/mcp/thumbnails/developer-tools.png",[827,829,831],{"title":430,"description":828},"Evaluated strictly for implementation of official @modelcontextprotocol/sdk or vendor-backed Go/Python MCP libraries with standard tool schemas.",{"title":433,"description":830},"Every callable tool name, input parameter, CLI launch string, and claude_desktop_config.json entry is verified against the actual repository source code.",{"title":436,"description":832},"Tested for context-window efficiency: avoiding unpruned raw payload dumps and enforcing structured JSON filtering before LLM consumption.","developer-tools","mcp/developer-tools","August 2026","5zqiEXGyIhjli0iuhs18xV61amHn6NDcZtRE3KPkv7s",{"id":838,"title":839,"category":840,"decisionMatrix":841,"extension":59,"faqs":882,"items":895,"meta":1227,"metaDescription":1228,"metaTitle":1229,"ogImage":1230,"principles":1231,"slug":1237,"stem":1238,"updatedAt":440,"__hash__":1239},"mcp/mcp/voice-agents.yml","Top 10 Voice AI & Telephony MCP Servers for AI Agents (2026)","Voice AI & Telephony",[842,846,850,854,858,862,866,870,874,878],{"step":10,"trigger":843,"recommendation":844,"reason":845},"You need the highest quality speech synthesis, voice cloning, and sound effects with full conversational agent creation","elevenlabs-mcp","The official vendor-maintained server with the most complete audio synthesis and voice design toolset.",{"step":15,"trigger":847,"recommendation":848,"reason":849},"You need programmable SMS messaging, telephone call dispatching, and carrier phone number management","twilio-mcp","Backed by official Twilio OpenAPI specs providing end-to-end access to Twilio communication APIs.",{"step":20,"trigger":851,"recommendation":852,"reason":853},"You require ultra-low latency streaming voice responses with accent adaptation and URL-based audio links","cartesia-mcp","Engineered specifically for low-latency streaming TTS, returning clean download URLs rather than context-heavy byte arrays.",{"step":25,"trigger":855,"recommendation":856,"reason":857},"You want to deploy, test, and dial full conversational voice agents configured with custom LLMs and transcribers","vapi-mcp","Offers end-to-end assistant configuration, phone number provisioning, and outbound call dispatching.",{"step":30,"trigger":859,"recommendation":860,"reason":861},"You use Retell AI to manage voice bots and need to create web call tokens or place telephone calls","retellai-mcp","Provides complete control plane coverage for Retell AI assistants, dynamic LLM prompts, and call analytics.",{"step":35,"trigger":863,"recommendation":864,"reason":865},"You want to convert research documents and web pages into two-speaker conversational podcast audio","notebooklm-mcp","Automates Google NotebookLM workflows to produce multi-speaker audio overviews and citation-backed summaries.",{"step":40,"trigger":867,"recommendation":868,"reason":869},"You need zero-cost, completely offline speech synthesis without API keys or cloud dependencies","kokoro-tts-mcp","Runs open-weight Kokoro-82M models locally with multi-speaker script and podcast generation tools.",{"step":45,"trigger":871,"recommendation":872,"reason":873},"You need asynchronous speech-to-text transcription with background job tracking so the MCP transport never hangs","stt-mcp","Features process isolation, non-blocking job status handles, and fail-closed GPU management.",{"step":50,"trigger":875,"recommendation":876,"reason":877},"You develop on Apple Silicon Macs and want fast, private on-device Whisper transcription","mlx-whisper-mcp","Hardware-accelerated for Apple M-series chips with zero configuration and local MLX execution.",{"step":55,"trigger":879,"recommendation":880,"reason":881},"You want community-driven Google NotebookLM document grounding and audio overview generation","pleaseprompto-notebooklm","The most starred community server for automating research queries and audio discussions.",[883,886,889,892],{"question":884,"answer":885},"How do voice AI MCP servers handle large audio payloads without exhausting LLM context tokens?","Servers take different approaches to audio data. Modern streaming servers like `cartesia-mcp` write audio to persistent storage and return short download URLs, keeping LLM prompts concise. In contrast, servers configured to return raw base64 audio strings in `elevenlabs-mcp` can quickly consume hundreds of thousands of tokens, making URL-based or file-based output modes strongly preferable for agent workflows.",{"question":887,"answer":888},"Can voice AI MCP servers place real telephone calls to customers?","Yes, servers like `twilio-mcp`, `vapi-mcp`, `retellai-mcp`, and `elevenlabs-mcp` include tools to initiate outbound PSTN telephone calls. Because these tools place live calls and incur carrier or platform charges immediately, teams should implement human-in-the-loop approval checkpoints before granting agents access to outbound dialing tools.",{"question":890,"answer":891},"Which voice and speech recognition MCP servers can run completely offline?","Both `kokoro-tts-mcp` for speech synthesis and `stt-mcp` or `mlx-whisper-mcp` for speech-to-text run completely on local hardware without sending data to cloud APIs. This ensures zero API costs and full data privacy for sensitive voice recordings.",{"question":893,"answer":894},"How does Webfuse integrate with Voice AI and Telephony MCP servers?","Webfuse provides live web session virtualization and zero-code co-browsing. When combined with voice MCP servers like `elevenlabs-mcp` or `vapi-mcp`, AI agents can speak with customers in real time while simultaneously seeing, navigating, and filling out live web interfaces with human approval gates.",[896,929,964,996,1030,1064,1098,1130,1163,1197],{"rank":10,"slug":844,"name":897,"author":898,"repo":897,"githubUrl":899,"stars":762,"transport":119,"ogImage":900,"installCommand":901,"bestFor":902,"boundary":903,"boundaryNote":904,"summary":905,"tools":906,"configSnippet":923,"limitations":924,"webfuseSynergy":927,"runtimes":928},"elevenlabs/elevenlabs-mcp","ElevenLabs (Official)","https://github.com/elevenlabs/elevenlabs-mcp","https://opengraph.githubassets.com/1/elevenlabs/elevenlabs-mcp","uvx elevenlabs-mcp","High-fidelity speech synthesis, voice cloning, audio isolation, and conversational voice agent orchestration","Speech Synthesis & Conversational AI","Connects agents directly to ElevenLabs voice generation, sound effects, speech-to-text, and conversational AI endpoints.","The official ElevenLabs Model Context Protocol server. Built on FastMCP and the official Python SDK, it exposes 27 tools covering text-to-speech, speech-to-text with diarization, sound effect generation, conversational agent creation, and outbound call simulation.",[907,911,915,919],{"name":908,"schema":909,"description":910},"text_to_speech","{ \"text\": \"string\", \"voice_name\": \"string?\", \"stability\": \"number?\", \"output_format\": \"string?\" }","Generate natural speech audio from text using ElevenLabs voice models.",{"name":912,"schema":913,"description":914},"speech_to_text","{ \"input_file_path\": \"string\", \"language_code\": \"string?\", \"diarize\": \"boolean?\" }","Transcribe audio files with speaker diarization.",{"name":916,"schema":917,"description":918},"create_agent","{ \"name\": \"string\", \"first_message\": \"string\", \"system_prompt\": \"string\", \"voice_id\": \"string?\" }","Deploy a conversational voice agent with system prompts and custom voices.",{"name":920,"schema":921,"description":922},"make_outbound_call","{ \"agent_id\": \"string\", \"agent_phone_number_id\": \"string\", \"to_number\": \"string\" }","Initiate an outbound phone call connecting a voice agent to a phone number.","{\n  \"mcpServers\": {\n    \"elevenlabs\": {\n      \"command\": \"uvx\",\n      \"args\": [\"elevenlabs-mcp\"],\n      \"env\": {\n        \"ELEVENLABS_API_KEY\": \"your-api-key\",\n        \"ELEVENLABS_MCP_OUTPUT_MODE\": \"files\"\n      }\n    }\n  }\n}\n",[925,926],"Setting output mode to resources returns base64 audio in the prompt context, increasing token consumption.","Outbound telephony calls and voice cloning tools lack dry-run validation flags.","Pair ElevenLabs voice synthesis with Webfuse live browser sessions to narrate and execute web actions in real time.",[109,110,111,112],{"rank":15,"slug":848,"name":930,"author":931,"repo":932,"githubUrl":933,"stars":934,"transport":119,"ogImage":935,"installCommand":936,"bestFor":937,"boundary":938,"boundaryNote":939,"summary":940,"tools":941,"configSnippet":958,"limitations":959,"webfuseSynergy":962,"runtimes":963},"@twilio-alpha/mcp","Twilio (Official Alpha)","twilio-labs/mcp","https://github.com/twilio-labs/mcp","109","https://opengraph.githubassets.com/1/twilio-labs/mcp","npx -y @twilio-alpha/mcp","Enterprise SMS messaging, voice call dispatching, and phone number administration via OpenAPI schemas","Telephony & SMS Communications","Translates Twilio OpenAPI specifications into callable MCP tools for programmable telephony.","The official Twilio Alpha MCP server. Written in TypeScript, it translates Twilio OpenAPI definitions into tools that let agents send SMS messages, initiate programmable voice calls, manage phone numbers, and inspect call logs.",[942,946,950,954],{"name":943,"schema":944,"description":945},"twilio_api_v2010--CreateMessage","{ \"To\": \"string\", \"From\": \"string?\", \"Body\": \"string?\" }","Send an outbound SMS message via the Twilio API.",{"name":947,"schema":948,"description":949},"twilio_api_v2010--CreateCall","{ \"To\": \"string\", \"From\": \"string\", \"Url\": \"string?\" }","Initiate an outbound programmable voice call connected to TwiML instructions.",{"name":951,"schema":952,"description":953},"twilio_api_v2010--ListMessage","{ \"To\": \"string?\", \"From\": \"string?\", \"PageSize\": \"number?\" }","Query sent and received SMS communication records.",{"name":955,"schema":956,"description":957},"twilio_api_v2010--ListCall","{ \"Status\": \"string?\", \"StartTime\": \"string?\", \"PageSize\": \"number?\" }","Retrieve voice call logs, durations, and completion statuses.","{\n  \"mcpServers\": {\n    \"twilio\": {\n      \"command\": \"npx\",\n      \"args\": [\n        \"-y\",\n        \"@twilio-alpha/mcp\",\n        \"-a\",\n        \"ACXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX\",\n        \"-k\",\n        \"SKXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX\",\n        \"-s\",\n        \"your-api-secret\",\n        \"--services\",\n        \"twilio_api_v2010\"\n      ]\n    }\n  }\n}\n",[960,961],"Loading default services registers hundreds of tools that consume significant context tokens without filtering.","Destructive actions and billable SMS calls execute without separate approval prompts.","Trigger Webfuse authenticated browsing sessions when a customer requests support through Twilio SMS.",[109,110,111],{"rank":20,"slug":852,"name":852,"author":965,"repo":966,"githubUrl":967,"stars":968,"transport":79,"ogImage":969,"installCommand":970,"bestFor":971,"boundary":972,"boundaryNote":973,"summary":974,"tools":975,"configSnippet":990,"limitations":991,"webfuseSynergy":994,"runtimes":995},"Cartesia (Official)","cartesia-ai/cartesia-mcp","https://github.com/cartesia-ai/cartesia-mcp","14","https://opengraph.githubassets.com/1/cartesia-ai/cartesia-mcp","uvx cartesia-mcp","Low-latency streaming speech synthesis, voice localization, accent control, and pronunciation management","Real-Time Streaming TTS","Wraps Cartesia Sonic models with structured rate limiting and audio file references.","The official Cartesia Model Context Protocol server. Designed for real-time voice applications, it supports dual stdio and Streamable HTTP transports, returning downloadable audio URLs instead of embedding raw audio bytes directly into context.",[976,979,983,987],{"name":908,"schema":977,"description":978},"{ \"transcript\": \"string\", \"voice_id\": \"string\", \"output_format\": \"object\", \"language\": \"string?\" }","Generate natural speech audio using Cartesia Sonic models.",{"name":980,"schema":981,"description":982},"localize_voice","{ \"voice_id\": \"string\", \"name\": \"string\", \"accent\": \"string\", \"language\": \"string\" }","Adapt existing voices to regional dialects and accents.",{"name":984,"schema":985,"description":986},"list_voices","{ \"limit\": \"number?\", \"gender\": \"string?\", \"language\": \"string?\" }","List available voices with metadata and pagination cursors.",{"name":912,"schema":988,"description":989},"{ \"file_path\": \"string\", \"mode\": \"string?\", \"language\": \"string?\" }","Transcribe audio files with word-level timestamp precision.","{\n  \"mcpServers\": {\n    \"cartesia\": {\n      \"command\": \"uvx\",\n      \"args\": [\"cartesia-mcp\"],\n      \"env\": {\n        \"CARTESIA_API_KEY\": \"your-cartesia-api-key\"\n      }\n    }\n  }\n}\n",[992,993],"Requires Python 3.13 or newer, which may require environment adjustments on older host runtimes.","Voice cloning and change operations require audio files stored on the local MCP server filesystem.","Deliver low-latency voice responses in Webfuse co-browsing sessions using Cartesia streaming audio.",[109,110,111],{"rank":25,"slug":856,"name":997,"author":998,"repo":999,"githubUrl":1000,"stars":1001,"transport":119,"ogImage":1002,"installCommand":1003,"bestFor":1004,"boundary":1005,"boundaryNote":1006,"summary":1007,"tools":1008,"configSnippet":1024,"limitations":1025,"webfuseSynergy":1028,"runtimes":1029},"@vapi-ai/mcp-server","Vapi AI (Official)","VapiAI/mcp-server","https://github.com/VapiAI/mcp-server","57","https://opengraph.githubassets.com/1/VapiAI/mcp-server","npx -y @vapi-ai/mcp-server","Configuring conversational voice assistants, managing transcribers, and triggering outbound voice calls","Voice Assistant Platform","Manages Vapi voice agents, phone numbers, customer call dispatch, and dynamic assistant tools.","The official Vapi Model Context Protocol server. Running on Node with the Vapi server SDK, it provides tools to create voice assistants, configure transcribers and speech models, provision phone numbers, and place outbound telephone calls.",[1009,1013,1017,1020],{"name":1010,"schema":1011,"description":1012},"create_assistant","{ \"name\": \"string\", \"instructions\": \"string?\", \"voice\": \"object?\", \"transcriber\": \"object?\" }","Create and configure a full conversational voice assistant.",{"name":1014,"schema":1015,"description":1016},"create_call","{ \"assistantId\": \"string?\", \"phoneNumberId\": \"string?\", \"customer\": \"object\" }","Dispatch an outbound phone call connecting a customer to a voice assistant.",{"name":1018,"schema":172,"description":1019},"list_assistants","List active voice assistants configured in your Vapi organization.",{"name":1021,"schema":1022,"description":1023},"get_call","{ \"callId\": \"string\" }","Retrieve call duration, status, cost, and execution details.","{\n  \"mcpServers\": {\n    \"vapi\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"@vapi-ai/mcp-server\"],\n      \"env\": {\n        \"VAPI_TOKEN\": \"your-vapi-token\"\n      }\n    }\n  }\n}\n",[1026,1027],"List tools use fixed result caps without cursor pagination for large assistant directories.","Placing phone calls incurs telephony costs immediately without a separate dry-run preview parameter.","Embed Vapi voice assistants inside Webfuse web sessions to let AI agents talk to customers while co-browsing.",[109,110,111],{"rank":30,"slug":860,"name":1031,"author":1032,"repo":1033,"githubUrl":1034,"stars":1035,"transport":119,"ogImage":1036,"installCommand":1037,"bestFor":1038,"boundary":1039,"boundaryNote":1040,"summary":1041,"tools":1042,"configSnippet":1058,"limitations":1059,"webfuseSynergy":1062,"runtimes":1063},"retellai-mcp-server","abhaybabbar","abhaybabbar/retellai-mcp-server","https://github.com/abhaybabbar/retellai-mcp-server","39","https://opengraph.githubassets.com/1/abhaybabbar/retellai-mcp-server","npx -y @abhaybabbar/retellai-mcp-server","Retell AI conversational agent management, dynamic LLM prompt configuration, and web call creation","Conversational Telephony Agents","Bridges Retell AI agent configurations, phone numbers, and call analytics to MCP clients.","A community TypeScript MCP server for Retell AI. It exposes 21 tools covering agent creation, dynamic variable injection, phone number provisioning, and initiating web-based and PSTN telephone calls.",[1043,1047,1051,1054],{"name":1044,"schema":1045,"description":1046},"create_phone_call","{ \"fromNumber\": \"string\", \"toNumber\": \"string\", \"overrideAgentId\": \"string?\" }","Place an outbound phone call using a configured Retell agent.",{"name":1048,"schema":1049,"description":1050},"create_web_call","{ \"agentId\": \"string\", \"metadata\": \"object?\" }","Generate a web call token for browser-based voice conversations.",{"name":916,"schema":1052,"description":1053},"{ \"llmWebsocketUrl\": \"string\", \"voiceId\": \"string\", \"agentName\": \"string?\" }","Register a new conversational voice agent with Retell.",{"name":1055,"schema":1056,"description":1057},"list_calls","{ \"agentId\": \"string?\", \"limit\": \"number?\", \"offset\": \"number?\" }","Query past call records, durations, and call analysis metrics.","{\n  \"mcpServers\": {\n    \"retell\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"@abhaybabbar/retellai-mcp-server\"],\n      \"env\": {\n        \"RETELL_API_KEY\": \"your-retell-api-key\"\n      }\n    }\n  }\n}\n",[1060,1061],"Resource deletion tools are exposed unconditionally without separate verification prompts.","Does not implement automatic request retries or backoff when hitting API limits.","Connect Retell web calls into Webfuse live web spaces for voice-guided website onboarding.",[109,110,111],{"rank":35,"slug":864,"name":864,"author":1065,"repo":1066,"githubUrl":1067,"stars":1068,"transport":1069,"ogImage":1070,"installCommand":1071,"bestFor":1072,"boundary":1073,"boundaryNote":1074,"summary":1075,"tools":1076,"configSnippet":1092,"limitations":1093,"webfuseSynergy":1096,"runtimes":1097},"Romain Peyrichou (roomi-fields)","roomi-fields/notebooklm-mcp","https://github.com/roomi-fields/notebooklm-mcp","170","stdio / HTTP REST","https://opengraph.githubassets.com/1/roomi-fields/notebooklm-mcp","npx -y @roomi-fields/notebooklm-mcp","Generating structured audio overviews, multi-speaker podcasts, and citation-backed summaries from source documents","Audio Overview & Podcast Generation","Drives Google NotebookLM workflows to convert research notebooks into multi-speaker conversational audio.","A feature-rich MCP server and REST service for Google NotebookLM. It connects research notes, document sources, and study aids to generate two-speaker podcast discussions, audio overviews, and structured study guides.",[1077,1081,1085,1089],{"name":1078,"schema":1079,"description":1080},"generate_content","{ \"content_type\": \"string\", \"custom_instructions\": \"string?\", \"language\": \"string?\" }","Generate conversational audio overviews, reports, or study aids from notebook sources.",{"name":1082,"schema":1083,"description":1084},"ask_question","{ \"question\": \"string\", \"notebook_id\": \"string?\" }","Query notebook documents with grounded, citation-backed answers.",{"name":1086,"schema":1087,"description":1088},"add_source","{ \"notebook_id\": \"string\", \"source_type\": \"string\", \"content\": \"string\" }","Add a document, web URL, or text note to a notebook.",{"name":1090,"schema":172,"description":1091},"list_notebooks","List accessible notebooks and active document collections.","{\n  \"mcpServers\": {\n    \"notebooklm\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"@roomi-fields/notebooklm-mcp\"],\n      \"env\": {\n        \"NOTEBOOKLM_CONTENT_LANGUAGE\": \"en\"\n      }\n    }\n  }\n}\n",[1094,1095],"Relies on browser automation and undocumented Google endpoints, requiring dedicated account cookies.","Subject to daily account generation quotas for long audio overviews.","Scrape reference documentation through Webfuse proxies and pass clean markdown into NotebookLM for audio podcast generation.",[109,110,111],{"rank":40,"slug":868,"name":1099,"author":1100,"repo":1101,"githubUrl":1102,"stars":968,"transport":119,"ogImage":1103,"installCommand":1104,"bestFor":1105,"boundary":1106,"boundaryNote":1107,"summary":1108,"tools":1109,"configSnippet":1124,"limitations":1125,"webfuseSynergy":1128,"runtimes":1129},"kokoro-mcp-server","Aparsoft","aparsoft/kokoro-mcp-server","https://github.com/aparsoft/kokoro-mcp-server","https://opengraph.githubassets.com/1/aparsoft/kokoro-mcp-server","pip install -e '.[mcp,cli]'","Completely offline, zero-cost text-to-speech synthesis and automated podcast generation using open weights","Open-Weights Local TTS","Runs local Kokoro-82M speech synthesis models with audio enhancement and script processing.","An open-source Python MCP server built on the Kokoro-82M model. It provides private, local speech generation without API keys or cloud dependencies, featuring audio enhancement, multi-speaker podcast generation, and silence trimming.",[1110,1114,1118,1120],{"name":1111,"schema":1112,"description":1113},"generate_speech","{ \"text\": \"string\", \"voice\": \"string?\", \"speed\": \"number?\", \"output_file\": \"string?\" }","Synthesize speech locally using Kokoro voice models.",{"name":1115,"schema":1116,"description":1117},"generate_podcast","{ \"segments\": \"array\", \"output_path\": \"string\", \"gap_duration\": \"number?\" }","Orchestrate multi-speaker podcast dialogues with customizable pacing.",{"name":984,"schema":172,"description":1119},"List available local voice models and language presets.",{"name":1121,"schema":1122,"description":1123},"process_script","{ \"script_path\": \"string\", \"output_path\": \"string\", \"voice\": \"string?\" }","Convert a multi-line text script into sequential audio files.","{\n  \"mcpServers\": {\n    \"kokoro\": {\n      \"command\": \"aparsoft-tts-server\",\n      \"args\": [],\n      \"env\": {\n        \"TTS_VOICE\": \"am_michael\",\n        \"TTS_SPEED\": \"1.0\"\n      }\n    }\n  }\n}\n",[1126,1127],"Downloads local model weights on first execution, introducing initial setup latency.","Synthesis speed depends on host CPU or GPU capabilities during batch audio generation.","Run zero-cost offline speech generation to provide local audio prompts for browser automation testing in Webfuse.",[109,110,111],{"rank":45,"slug":872,"name":872,"author":1131,"repo":1132,"githubUrl":1133,"stars":1134,"transport":119,"ogImage":1135,"installCommand":1136,"bestFor":1137,"boundary":1138,"boundaryNote":1139,"summary":1140,"tools":1141,"configSnippet":1157,"limitations":1158,"webfuseSynergy":1161,"runtimes":1162},"sm18lr88","sm18lr88/STT-MCP","https://github.com/sm18lr88/STT-MCP","31","https://opengraph.githubassets.com/1/sm18lr88/STT-MCP","uv run stt-mcp","Local, private speech-to-text transcription with asynchronous job lifecycle management and CUDA safety policies","Local Speech Recognition & Async Jobs","Provides process-isolated speech-to-text powered by Granite Speech or parakeet.cpp.","A local speech-to-text MCP server designed for privacy and stability. It features process-isolated workers, fail-closed GPU policies, and asynchronous job tracking to prevent blocking MCP transports during long audio transcriptions.",[1142,1146,1150,1154],{"name":1143,"schema":1144,"description":1145},"start_transcription","{ \"job_id\": \"string\", \"source_path\": \"string\", \"formats\": \"array?\" }","Start an asynchronous transcription job returning an immediate status handle.",{"name":1147,"schema":1148,"description":1149},"get_transcription_status","{ \"job_id\": \"string\" }","Check progress, execution state, and output paths of a transcription job.",{"name":1151,"schema":1152,"description":1153},"transcribe","{ \"source_path\": \"string\", \"formats\": \"array?\" }","Transcribe an audio file synchronously and return structured text results.",{"name":1155,"schema":1148,"description":1156},"cancel_transcription","Cancel an ongoing background transcription task.","{\n  \"mcpServers\": {\n    \"stt\": {\n      \"command\": \"uv\",\n      \"args\": [\"run\", \"stt-mcp\"]\n    }\n  }\n}\n",[1159,1160],"Operates one transcription job at a time per worker process, rejecting concurrent requests as busy.","Requires local audio file paths on the host machine rather than remote URLs or base64 streams.","Transcribe customer audio feedback locally and link transcripts to Webfuse session diagnostic reports.",[109,110,111],{"rank":50,"slug":876,"name":876,"author":1164,"repo":1165,"githubUrl":1166,"stars":1167,"transport":119,"ogImage":1168,"installCommand":1169,"bestFor":1170,"boundary":1171,"boundaryNote":1172,"summary":1173,"tools":1174,"configSnippet":1191,"limitations":1192,"webfuseSynergy":1195,"runtimes":1196},"Kachi Odoemene","kachiO/mlx-whisper-mcp","https://github.com/kachiO/mlx-whisper-mcp","24","https://opengraph.githubassets.com/1/kachiO/mlx-whisper-mcp","uv run mlx_whisper_mcp.py","On-device speech-to-text transcription optimized specifically for Apple Silicon hardware","Apple Silicon STT","Uses Apple MLX frameworks for fast on-device Whisper Large v3 Turbo transcription.","A lightweight Python MCP server using Apple MLX Whisper. It enables fast on-device speech transcription and translation on macOS without cloud API keys, processing local files and base64 audio inputs.",[1175,1179,1183,1187],{"name":1176,"schema":1177,"description":1178},"transcribe_file","{ \"file_path\": \"string\", \"language\": \"string?\", \"task\": \"string?\" }","Transcribe an audio file into text using on-device MLX Whisper models.",{"name":1180,"schema":1181,"description":1182},"transcribe_audio","{ \"audio_data\": \"string\", \"file_format\": \"string\", \"language\": \"string?\" }","Transcribe base64-encoded audio data directly in memory.",{"name":1184,"schema":1185,"description":1186},"transcribe_youtube","{ \"url\": \"string\", \"language\": \"string?\", \"task\": \"string?\" }","Fetch and transcribe audio tracks from YouTube URLs.",{"name":1188,"schema":1189,"description":1190},"download_youtube","{ \"url\": \"string\", \"keep_file\": \"boolean?\" }","Download audio streams from video URLs to a local data directory.","{\n  \"mcpServers\": {\n    \"mlx-whisper\": {\n      \"command\": \"uv\",\n      \"args\": [\"run\", \"/path/to/mlx_whisper_mcp.py\"]\n    }\n  }\n}\n",[1193,1194],"Compatible only with Apple Silicon (M-series) Macs running macOS.","Initial execution downloads a 1.6GB Whisper Large v3 model before transcription begins.","Run private, on-device audio transcription on Mac developer machines while testing Webfuse voice agents.",[109,110,111],{"rank":55,"slug":880,"name":1198,"author":1199,"repo":1200,"githubUrl":1201,"stars":1202,"transport":119,"ogImage":1203,"installCommand":1204,"bestFor":1205,"boundary":1206,"boundaryNote":1207,"summary":1208,"tools":1209,"configSnippet":1221,"limitations":1222,"webfuseSynergy":1225,"runtimes":1226},"notebooklm-mcp (PleasePrompto)","PleasePrompto","PleasePrompto/notebooklm-mcp","https://github.com/PleasePrompto/notebooklm-mcp","3.4k","https://opengraph.githubassets.com/1/PleasePrompto/notebooklm-mcp","npx -y notebooklm-mcp","Document research, citation extraction, and triggering Google NotebookLM deep research audio overviews","Research & Audio Summaries","Connects agents to NotebookLM chat sessions and conversational audio generation tools.","The widely adopted community MCP server for Google NotebookLM. It allows coding agents to query source notebooks, manage document attachments, and initiate deep conversational audio overviews.",[1210,1212,1215,1219],{"name":1082,"schema":1083,"description":1211},"Ask questions grounded in notebook sources with exact citations.",{"name":1086,"schema":1213,"description":1214},"{ \"notebook_id\": \"string\", \"url\": \"string?\", \"text\": \"string?\" }","Add a web link, text document, or PDF to a research notebook.",{"name":1216,"schema":1217,"description":1218},"generate_audio_overview","{ \"notebook_id\": \"string\", \"instructions\": \"string?\" }","Trigger a two-speaker conversational audio summary of the notebook sources.",{"name":1090,"schema":172,"description":1220},"List all active notebooks and document counts.","{\n  \"mcpServers\": {\n    \"notebooklm-prompto\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"notebooklm-mcp\"]\n    }\n  }\n}\n",[1223,1224],"Requires persistent browser cookies and manual Google login setup.","Long audio generation tasks can take several minutes to process on Google servers.","Gather documentation from protected web pages using Webfuse stealth proxies before submitting to NotebookLM.",[109,110,111],{},"The top 10 Model Context Protocol (MCP) servers for ElevenLabs, Twilio telephony, Cartesia streaming TTS, Vapi, Retell AI, and local Whisper speech recognition, with tool schemas and configs for Claude Desktop, Cursor, and OpenCode.","Top 10 Voice AI & Telephony MCP Servers for AI Agents (2026) | Webfuse","/mcp/thumbnails/voice-agents.png",[1232,1234,1235],{"title":430,"description":1233},"Evaluated for implementation of standard MCP transports over stdio and Streamable HTTP, supporting tool schemas, resources, and audio generation endpoints.",{"title":433,"description":434},{"title":436,"description":1236},"Assessed on how responses handle large audio payloads, transcript streaming, and base64 payloads to prevent model context exhaustion.","voice-agents","mcp/voice-agents","SqcVn_OJiZBFx5MtGjcN8giGa9XaZPT-JcyNASXxfx4",{"id":1241,"title":1242,"category":1243,"decisionMatrix":1244,"extension":59,"faqs":1285,"items":1298,"meta":1617,"metaDescription":1618,"metaTitle":1619,"ogImage":1620,"principles":1621,"slug":1628,"stem":1629,"updatedAt":835,"__hash__":1630},"mcp/mcp/web-scraping.yml","Top 10 Web Scraping & Content Extraction MCP Servers for AI Agents (2026)","Web Scraping & Data Extraction",[1245,1249,1253,1257,1261,1265,1269,1273,1277,1281],{"step":10,"trigger":1246,"recommendation":1247,"reason":1248},"Need to crawl entire websites, follow links, and convert full pages to LLM-ready markdown","firecrawl-mcp","Leading crawler server with asynchronous job queues, sitemap parsing, and structured JSON extraction.",{"step":15,"trigger":1250,"recommendation":1251,"reason":1252},"Require a high-speed, free, self-hosted crawler with local browser pooling and heuristic extraction","crawl4ai-mcp","Open-source async engine with deep crawling algorithms and zero per-page API fees.",{"step":20,"trigger":1254,"recommendation":1255,"reason":1256},"Target site uses Cloudflare Turnstile bot challenges or contains hidden prompt injections","scrapling-mcp","Features automated challenge handling and prompt injection stripping from untrusted HTML.",{"step":25,"trigger":1258,"recommendation":1259,"reason":1260},"Searching for code, company data, or technical documentation via semantic embeddings","exa-mcp","Neural search engine built specifically for AI agents to retrieve high-signal web context.",{"step":30,"trigger":1262,"recommendation":1263,"reason":1264},"Need real-time search grounding and concise research summaries in one step","tavily-mcp","Purpose-built for agent search grounding with pre-digested answers and extraction tools.",{"step":35,"trigger":1266,"recommendation":1267,"reason":1268},"Need structured data from Amazon, LinkedIn, Google Maps, or social platforms without writing custom scrapers","apify-mcp","Provides access to over 60,000 managed cloud scrapers with pre-configured schemas.",{"step":40,"trigger":1270,"recommendation":1271,"reason":1272},"Want independent web search with local POI lookups and no user tracking","brave-search-mcp","Official Brave server offering global web indexing and local place search with DNS protection.",{"step":45,"trigger":1274,"recommendation":1275,"reason":1276},"Need clean markdown extraction from URLs or academic papers without requiring an API key","jina-ai-mcp","High-quality Jina Reader integration with listwise neural reranking on search queries.",{"step":50,"trigger":1278,"recommendation":1279,"reason":1280},"Encountering hard geographic IP blocks or enterprise CAPTCHA challenges","brightdata-mcp","Enterprise unblocker with global residential proxy networks and scraping browser sessions.",{"step":55,"trigger":1282,"recommendation":1283,"reason":1284},"Need lightweight, keyless URL reading with strict robots.txt adherence","mcp-server-fetch","Official minimal reference implementation with character pagination and markdown conversion.",[1286,1289,1292,1295],{"question":1287,"answer":1288},"What is the difference between a web scraping MCP and a browser automation MCP?","Web scraping MCP servers (like Firecrawl, Crawl4AI, and Jina) focus on retrieving and converting web content into clean Markdown or structured JSON for LLM context. Browser automation MCP servers (like Playwright and Puppeteer) maintain interactive browser sessions that can click buttons, fill out forms, take screenshots, and execute client-side scripts.",{"question":1290,"answer":1291},"How do scraping MCP servers handle Cloudflare bot detection and CAPTCHAs?","Basic fetch tools will be blocked by Cloudflare, Akamai, or DataDome bot challenges. Advanced servers like Scrapling, Crawl4AI, and Bright Data use stealth browser engines, residential proxy rotation, and automated challenge solving to access protected content.",{"question":1293,"answer":1294},"Why do some servers require Docker instead of npx or uvx?","Servers like Crawl4AI and Bright Data run full Chromium browser instances or Python microservice bridges that have system-level dependencies. Running them inside Docker containers isolates browser processes and ensures consistent execution across macOS, Linux, and Windows.",{"question":1296,"answer":1297},"Can MCP web scraping tools cause token context window overflow?","Yes. Raw HTML pages often exceed 50,000 to 100,000 tokens of boilerplate tags, script payloads, and styles. All recommended servers in this guide clean and convert HTML to dense Markdown or extract specific JSON fields to preserve agent context windows.",[1299,1333,1364,1394,1424,1458,1492,1527,1561,1594],{"rank":10,"slug":1247,"name":1300,"author":1301,"repo":1300,"githubUrl":1302,"stars":1303,"transport":544,"ogImage":1304,"installCommand":1305,"bestFor":1306,"boundary":1307,"boundaryNote":1308,"summary":1309,"tools":1310,"configSnippet":1327,"limitations":1328,"webfuseSynergy":1331,"runtimes":1332},"firecrawl/firecrawl-mcp-server","Firecrawl (Mendable)","https://github.com/firecrawl/firecrawl-mcp-server","7.4k","https://opengraph.githubassets.com/1/firecrawl/firecrawl-mcp-server","npx -y firecrawl-mcp","Full-page scraping, multi-page site crawling, and structured JSON extraction from complex web targets","LLM-Ready Scraping & Crawling","Turns dynamic JS-heavy websites and sitemaps into clean, formatted Markdown and structured schemas.","The official MCP server from Firecrawl. It allows AI agents to scrape individual URLs, map site architectures from sitemaps, crawl entire domains asynchronously, and extract structured JSON schemas using LLM-driven parsing.",[1311,1315,1319,1323],{"name":1312,"schema":1313,"description":1314},"firecrawl_scrape","{ \"url\": \"string\", \"search\": \"string?\", \"sitemap\": \"string?\", \"includeSubdomains\": \"boolean?\" }","Scrape a single webpage and return clean markdown content.",{"name":1316,"schema":1317,"description":1318},"firecrawl_crawl","{ \"url\": \"string\", \"limit\": \"number?\", \"scrapeOptions\": \"object?\" }","Initiate an asynchronous crawl job across an entire domain or path.",{"name":1320,"schema":1321,"description":1322},"firecrawl_map","{ \"url\": \"string\", \"search\": \"string?\", \"limit\": \"number?\" }","Map all discovered URLs on a website to plan targeted scraping.",{"name":1324,"schema":1325,"description":1326},"firecrawl_extract","{ \"urls\": \"string[]\", \"prompt\": \"string?\", \"schema\": \"object?\" }","Extract structured JSON entities from one or multiple URLs using an extraction schema.","{\n  \"mcpServers\": {\n    \"firecrawl\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"firecrawl-mcp\"],\n      \"env\": {\n        \"FIRECRAWL_API_KEY\": \"fc-yourApiKeyHere\"\n      }\n    }\n  }\n}\n",[1329,1330],"Requires a Firecrawl API key or a self-hosted Firecrawl cluster instance.","Large recursive crawl jobs require polling crawl status rather than instant synchronous returns.","Route complex crawling jobs through Webfuse residential proxies to prevent rate limiting on strict commercial domains.",[109,110,111,112],{"rank":15,"slug":1251,"name":1334,"author":1335,"repo":1334,"githubUrl":1336,"stars":1337,"transport":1338,"ogImage":1339,"installCommand":1340,"bestFor":1341,"boundary":1342,"boundaryNote":1343,"summary":1344,"tools":1345,"configSnippet":1358,"limitations":1359,"webfuseSynergy":1362,"runtimes":1363},"unclecode/crawl4ai","UncleCode","https://github.com/unclecode/crawl4ai","80.6k","HTTP SSE / Docker Bridge","https://opengraph.githubassets.com/1/unclecode/crawl4ai","docker run -p 11235:11235 unclecode/crawl4ai","High-throughput self-hosted web crawling with anti-bot bypass and heuristic markdown filtering","Self-Hosted Async Crawling","Runs high-speed asynchronous browser pools locally with built-in bot detection countermeasures.","Crawl4AI is an open-source, high-speed asynchronous web crawler optimized for LLMs. It features deep crawling strategies (BFS/DFS), smart content chunking, media extraction, and automated bypass for common anti-bot mechanisms.",[1346,1350,1354],{"name":1347,"schema":1348,"description":1349},"crawl","{ \"url\": \"string\", \"word_count_threshold\": \"number?\", \"css_selector\": \"string?\", \"bypass_cache\": \"boolean?\" }","Crawl a target URL and return structured markdown with stripped boilerplates.",{"name":1351,"schema":1352,"description":1353},"deep_crawl","{ \"root_url\": \"string\", \"max_depth\": \"number?\", \"strategy\": \"string?\" }","Execute a multi-level recursive crawl across subpaths with configurable depth filters.",{"name":1355,"schema":1356,"description":1357},"extract_structured","{ \"url\": \"string\", \"schema\": \"object\", \"extraction_type\": \"string\" }","Extract typed data fields using CSS selectors, XPath rules, or LLM-based schemas.","{\n  \"mcpServers\": {\n    \"crawl4ai\": {\n      \"type\": \"http\",\n      \"url\": \"http://localhost:11235/mcp/sse\"\n    }\n  }\n}\n",[1360,1361],"Running local Chromium instances consumes notable memory and CPU on smaller host machines.","The native stdio transport has known edge cases; running via Docker SSE bridge is the recommended deployment.","Pair Crawl4AI with Webfuse proxy sessions to distribute request volume across multiple IP subnets.",[109,110,111],{"rank":20,"slug":1255,"name":1365,"author":1366,"repo":1365,"githubUrl":1367,"stars":1368,"transport":79,"ogImage":1369,"installCommand":1370,"bestFor":1371,"boundary":1372,"boundaryNote":1373,"summary":1374,"tools":1375,"configSnippet":1388,"limitations":1389,"webfuseSynergy":1392,"runtimes":1393},"D4Vinci/Scrapling","Karim Shoair (D4Vinci)","https://github.com/D4Vinci/Scrapling","77.5k","https://opengraph.githubassets.com/1/D4Vinci/Scrapling","pip install \"scrapling[ai]\" && scrapling mcp","Stealth web scraping that navigates Cloudflare Turnstile challenges and strips prompt injection","Stealth Scraping & Defense","Focuses on browser fingerprint masking, challenge handling, and sanitize-filtered extraction.","Scrapling is an adaptive Python scraping framework designed specifically for AI workflows. Its built-in MCP server exposes automated Cloudflare challenge solving, fingerprint rotation, adaptive element relocation, and automatic stripping of hidden prompt injection payloads from untrusted web HTML.",[1376,1380,1384],{"name":1377,"schema":1378,"description":1379},"fetch_page","{ \"url\": \"string\", \"mode\": \"string?\", \"wait_selector\": \"string?\" }","Fetch web pages using static, dynamic, or stealth Chrome engine modes.",{"name":1381,"schema":1382,"description":1383},"extract_markdown","{ \"url\": \"string\", \"main_content_only\": \"boolean?\", \"css_selector\": \"string?\" }","Extract clean Markdown scoped to main text elements with prompt injection tokens removed.",{"name":1385,"schema":1386,"description":1387},"solve_challenge","{ \"url\": \"string\", \"timeout\": \"number?\" }","Navigate to challenge-gated URLs and return verified session cookies.","{\n  \"mcpServers\": {\n    \"scrapling\": {\n      \"command\": \"scrapling\",\n      \"args\": [\"mcp\", \"--host\", \"127.0.0.1\", \"--port\", \"8000\", \"--no-auth\"]\n    }\n  }\n}\n",[1390,1391],"Requires local Chrome/Chromium installation for stealth dynamic fetch modes.","Advanced challenge resolution can add several seconds of latency per initial request.","Combine Scrapling's anti-injection sanitization with Webfuse session checkpoints for high-risk web workflows.",[109,110,111],{"rank":25,"slug":1259,"name":1395,"author":1396,"repo":1395,"githubUrl":1397,"stars":1398,"transport":79,"ogImage":1399,"installCommand":1400,"bestFor":1401,"boundary":1402,"boundaryNote":1403,"summary":1404,"tools":1405,"configSnippet":1418,"limitations":1419,"webfuseSynergy":1422,"runtimes":1423},"exa-labs/exa-mcp-server","Exa Labs","https://github.com/exa-labs/exa-mcp-server","4.9k","https://opengraph.githubassets.com/1/exa-labs/exa-mcp-server","npx -y exa-mcp-server","Neural semantic web search, AI-filtered link discovery, and structured research context","Neural Search & Web Fetch","Searches the web via semantic embeddings rather than keyword matching to return high-relevance pages.","The official MCP server from Exa (formerly Metaphor). Rather than standard keyword matching, Exa indexes the web using embeddings, allowing coding agents to find code repositories, developer documentation, company research, and live article text using natural language queries.",[1406,1410,1414],{"name":1407,"schema":1408,"description":1409},"web_search_exa","{ \"query\": \"string\", \"numResults\": \"number?\", \"includeDomains\": \"string[]?\", \"excludeDomains\": \"string[]?\" }","Perform semantic web search optimized for AI agent knowledge retrieval.",{"name":1411,"schema":1412,"description":1413},"web_fetch_exa","{ \"ids\": \"string[]\", \"text\": \"boolean?\", \"highlights\": \"boolean?\" }","Retrieve parsed page text and highlighted summary snippets from search result IDs.",{"name":1415,"schema":1416,"description":1417},"web_search_advanced_exa","{ \"query\": \"string\", \"startPublishedDate\": \"string?\", \"endPublishedDate\": \"string?\", \"category\": \"string?\" }","Execute structured search filtered by publishing date ranges and content categories.","{\n  \"mcpServers\": {\n    \"exa\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"exa-mcp-server\"],\n      \"env\": {\n        \"EXA_API_KEY\": \"your-exa-api-key\"\n      }\n    }\n  }\n}\n",[1420,1421],"Requires an Exa API key for high-volume or commercial usage.","Focuses on index-level content retrieval rather than direct real-time interactive DOM clicking.","Use Exa to find authoritative URLs across the web, then pass targeted pages to Webfuse browser agents for execution.",[109,110,111,112],{"rank":30,"slug":1263,"name":1425,"author":1426,"repo":1425,"githubUrl":1427,"stars":1428,"transport":79,"ogImage":1429,"installCommand":1430,"bestFor":1431,"boundary":1432,"boundaryNote":1433,"summary":1434,"tools":1435,"configSnippet":1452,"limitations":1453,"webfuseSynergy":1456,"runtimes":1457},"tavily-ai/tavily-mcp","Tavily AI","https://github.com/tavily-ai/tavily-mcp","2.4k","https://opengraph.githubassets.com/1/tavily-ai/tavily-mcp","npx -y tavily-mcp","Real-time search grounding, factual context extraction, and comprehensive research summaries","Real-Time Agent Search","Combines real-time web search and content parsing into pre-digested answers for LLMs.","Tavily's official MCP server provides an all-in-one search and extraction pipeline built for agents. It handles real-time search, raw page markdown extraction, domain mapping, and topic research without requiring separate scraping steps.",[1436,1440,1444,1448],{"name":1437,"schema":1438,"description":1439},"tavily_search","{ \"query\": \"string\", \"search_depth\": \"string?\", \"include_answer\": \"boolean?\", \"max_results\": \"number?\" }","Search the web for current information and return synthesized snippets.",{"name":1441,"schema":1442,"description":1443},"tavily_extract","{ \"urls\": \"string[]\", \"extract_depth\": \"string?\" }","Extract clean raw markdown text from a list of specific web page URLs.",{"name":1445,"schema":1446,"description":1447},"tavily_crawl","{ \"url\": \"string\", \"max_depth\": \"number?\", \"limit\": \"number?\" }","Crawl a website starting from a root URL and extract text content from discovered pages.",{"name":1449,"schema":1450,"description":1451},"tavily_map","{ \"url\": \"string\", \"max_depth\": \"number?\" }","Generate a site map outlining the structure and reachable links of a website.","{\n  \"mcpServers\": {\n    \"tavily\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"tavily-mcp\"],\n      \"env\": {\n        \"TAVILY_API_KEY\": \"tvly-yourApiKeyHere\"\n      }\n    }\n  }\n}\n",[1454,1455],"Monthly search quotas apply on free tiers.","Extract tool focuses on article content and drops interactive web application UI states.","Ground agent research in real-time facts with Tavily before triggering automated web interactions.",[109,110,111],{"rank":35,"slug":1267,"name":1459,"author":1460,"repo":1459,"githubUrl":1461,"stars":1462,"transport":79,"ogImage":1463,"installCommand":1464,"bestFor":1465,"boundary":1466,"boundaryNote":1467,"summary":1468,"tools":1469,"configSnippet":1486,"limitations":1487,"webfuseSynergy":1490,"runtimes":1491},"apify/apify-mcp-server","Apify","https://github.com/apify/apify-mcp-server","5.5k","https://opengraph.githubassets.com/1/apify/apify-mcp-server","npx -y @apify/actors-mcp-server","Executing pre-built cloud scrapers for social platforms, e-commerce, and Google SERPs at scale","Cloud Actor Ecosystem","Connects agents to thousands of specialized scrapers running in distributed cloud containers.","The official Apify MCP server bridges agents to the entire Apify Store ecosystem. Instead of writing custom scrapers for hard targets like Amazon, LinkedIn, Google Maps, or TikTok, agents search for, configure, and run purpose-built scrapers on cloud infrastructure.",[1470,1474,1478,1482],{"name":1471,"schema":1472,"description":1473},"search-actors","{ \"search\": \"string\", \"limit\": \"number?\" }","Find pre-built scrapers and actors across the Apify store catalog.",{"name":1475,"schema":1476,"description":1477},"call-actor","{ \"actorId\": \"string\", \"input\": \"object\", \"memoryMbytes\": \"number?\" }","Run a selected cloud scraper actor with custom parameters and retrieve the dataset.",{"name":1479,"schema":1480,"description":1481},"get-dataset-items","{ \"datasetId\": \"string\", \"limit\": \"number?\", \"offset\": \"number?\" }","Fetch structured JSON results from a completed actor crawl run.",{"name":1483,"schema":1484,"description":1485},"fetch-actor-details","{ \"actorId\": \"string\" }","Inspect the required input schema and documentation for any Apify actor.","{\n  \"mcpServers\": {\n    \"apify\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"@apify/actors-mcp-server\"],\n      \"env\": {\n        \"APIFY_TOKEN\": \"apify_api_yourTokenHere\"\n      }\n    }\n  }\n}\n",[1488,1489],"Actor runs consume Apify platform compute credits.","Large dataset exports require paginated item fetching to avoid context overflow.","Use Apify for large-scale structured data extraction and Webfuse for real-time live site navigation.",[109,110,111],{"rank":40,"slug":1271,"name":1493,"author":1494,"repo":1493,"githubUrl":1495,"stars":1496,"transport":1497,"ogImage":1498,"installCommand":1499,"bestFor":1500,"boundary":1501,"boundaryNote":1502,"summary":1503,"tools":1504,"configSnippet":1521,"limitations":1522,"webfuseSynergy":1525,"runtimes":1526},"brave/brave-search-mcp-server","Brave Software","https://github.com/brave/brave-search-mcp-server","1.4k","stdio (default) / HTTP","https://opengraph.githubassets.com/1/brave/brave-search-mcp-server","npx -y @brave/brave-search-mcp-server","Privacy-focused web search, local business lookups, and AI summary extraction","Independent Search Index","Provides an independent web search index free of third-party tracking algorithms.","The official Brave Search MCP server. Built on Brave's independent web index, it gives AI coding agents access to global web search, local points of interest, news articles, video metadata, and LLM-ready summarizer endpoints with DNS-rebinding security.",[1505,1509,1513,1517],{"name":1506,"schema":1507,"description":1508},"brave_web_search","{ \"query\": \"string\", \"count\": \"number?\", \"country\": \"string?\", \"search_lang\": \"string?\" }","Execute web search across Brave's independent index with language and region filters.",{"name":1510,"schema":1511,"description":1512},"brave_local_search","{ \"query\": \"string\", \"count\": \"number?\" }","Find local businesses, addresses, opening hours, and location reviews.",{"name":1514,"schema":1515,"description":1516},"brave_news_search","{ \"query\": \"string\", \"count\": \"number?\", \"freshness\": \"string?\" }","Search current news articles sorted by publication recency.",{"name":1518,"schema":1519,"description":1520},"brave_summarizer","{ \"key\": \"string\" }","Fetch pre-computed AI summaries generated directly by Brave Search algorithms.","{\n  \"mcpServers\": {\n    \"brave-search\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"@brave/brave-search-mcp-server\"],\n      \"env\": {\n        \"BRAVE_API_KEY\": \"BSA_yourApiKeyHere\"\n      }\n    }\n  }\n}\n",[1523,1524],"Requires a Brave Search API key (free tier includes 2,000 queries per month).","Returns search metadata and snippets rather than full DOM elements.","Use Brave Search for discovering target URLs and Webfuse for session-based authenticated site automation.",[109,110,111],{"rank":45,"slug":1275,"name":1528,"author":1529,"repo":1528,"githubUrl":1530,"stars":1531,"transport":1532,"ogImage":1533,"installCommand":1534,"bestFor":1535,"boundary":1536,"boundaryNote":1537,"summary":1538,"tools":1539,"configSnippet":1555,"limitations":1556,"webfuseSynergy":1559,"runtimes":1560},"jina-ai/MCP","Jina AI","https://github.com/jina-ai/MCP","0.8k","Streamable HTTP Remote / stdio","https://opengraph.githubassets.com/1/jina-ai/MCP","npx mcp-remote https://mcp.jina.ai/v1","Converting any URL into clean Markdown, deep reranked web search, and academic paper extraction","Markdown Reader & Reranking","Specializes in stripping web clutter, extracting clean text, and neural reranking.","The official MCP server from Jina AI. It provides instant URL-to-Markdown conversion through Jina Reader, deep web search with listwise neural reranking, PDF extraction, screenshot capture, and arXiv/SSRN paper discovery.",[1540,1544,1548,1552],{"name":1541,"schema":1542,"description":1543},"read_url","{ \"url\": \"string\", \"with_images_summary\": \"boolean?\", \"with_links_summary\": \"boolean?\" }","Fetch any webpage and return cleaned, LLM-ready Markdown text.",{"name":1545,"schema":1546,"description":1547},"search_web_deep","{ \"query\": \"string\", \"num\": \"number?\" }","Execute web search with listwise neural reranking applied to result passages.",{"name":1549,"schema":1550,"description":1551},"extract_pdf","{ \"url\": \"string\" }","Extract structured text and tables from online PDF documents.",{"name":1553,"schema":1550,"description":1554},"capture_screenshot_url","Capture a full-page screenshot of a target URL.","{\n  \"mcpServers\": {\n    \"jina\": {\n      \"type\": \"http\",\n      \"url\": \"https://mcp.jina.ai/v1\",\n      \"headers\": {\n        \"Authorization\": \"Bearer jina_yourApiKeyHere\"\n      }\n    }\n  }\n}\n",[1557,1558],"Keyless usage is rate-limited on heavy batch reading.","Dynamic Single Page Applications (SPAs) with complex interactions require headless browser engines.","Run Jina Reader through Webfuse to bypass regional IP blocks and access geo-restricted content.",[109,110,111],{"rank":50,"slug":1279,"name":1562,"author":1563,"repo":1562,"githubUrl":1564,"stars":1565,"transport":1566,"ogImage":1567,"installCommand":1568,"bestFor":1569,"boundary":1570,"boundaryNote":1571,"summary":1572,"tools":1573,"configSnippet":1588,"limitations":1589,"webfuseSynergy":1592,"runtimes":1593},"brightdata/brightdata-mcp","Bright Data","https://github.com/brightdata/brightdata-mcp","2.6k","stdio / Docker","https://opengraph.githubassets.com/1/brightdata/brightdata-mcp","docker run -i --rm -e API_TOKEN brightdata/ai-mcp","Enterprise unblocking, residential proxy rotation, and structured e-commerce extraction","Enterprise Unblocking & Proxies","Leverages a global residential proxy network to bypass strict bot defenses and CAPTCHAs.","Bright Data's official MCP server provides access to its Web Unlocker and Scraping Browser infrastructure. AI agents can execute multi-engine searches, scrape JS-rendered pages as markdown, manage headless browser sessions, and extract pre-indexed e-commerce datasets.",[1574,1577,1581,1584],{"name":1575,"schema":1550,"description":1576},"scrape_as_markdown","Scrape any URL via Web Unlocker and return clean Markdown.",{"name":1578,"schema":1579,"description":1580},"search_engine","{ \"query\": \"string\", \"engine\": \"string?\" }","Query Google, Bing, or Yandex with geographic proxy targeting.",{"name":1582,"schema":1550,"description":1583},"scraping_browser_navigate","Open and navigate a remote cloud browser session with automated CAPTCHA solving.",{"name":1585,"schema":1586,"description":1587},"scraping_browser_snapshot","{ \"full_page\": \"boolean?\" }","Capture visual screenshots and accessibility snapshots from the active cloud browser.","{\n  \"mcpServers\": {\n    \"brightdata\": {\n      \"command\": \"docker\",\n      \"args\": [\n        \"run\",\n        \"-i\",\n        \"--rm\",\n        \"-e\",\n        \"API_TOKEN\",\n        \"brightdata/ai-mcp\"\n      ],\n      \"env\": {\n        \"API_TOKEN\": \"your-brightdata-token\"\n      }\n    }\n  }\n}\n",[1590,1591],"Requires a paid Bright Data account and active zone credentials.","Docker execution requires a local Docker daemon running on the host.","Complement Bright Data extraction with Webfuse local session recording and human-in-the-loop approvals.",[109,110,111],{"rank":55,"slug":1283,"name":1595,"author":1596,"repo":1597,"githubUrl":1598,"stars":1599,"transport":572,"ogImage":1600,"installCommand":1601,"bestFor":1602,"boundary":1603,"boundaryNote":1604,"summary":1605,"tools":1606,"configSnippet":1611,"limitations":1612,"webfuseSynergy":1615,"runtimes":1616},"modelcontextprotocol/servers (fetch)","Model Context Protocol Core","modelcontextprotocol/servers","https://github.com/modelcontextprotocol/servers/tree/main/src/fetch","90.0k","https://opengraph.githubassets.com/1/modelcontextprotocol/servers","uvx --with \"mcp\u003C2\" mcp-server-fetch","Zero-configuration, keyless URL-to-Markdown reading with strict robots.txt compliance","Minimalist Reference Fetcher","A lightweight reference implementation for single-URL content retrieval without external API keys.","The official reference fetch server maintained by the Model Context Protocol core team. Written in Python, it converts web pages to clean Markdown using Readabilipy, respects site robots.txt rules, and supports chunked reading via character pagination.",[1607],{"name":1608,"schema":1609,"description":1610},"fetch","{ \"url\": \"string\", \"max_length\": \"number?\", \"start_index\": \"number?\", \"raw\": \"boolean?\" }","Fetch a web page and convert its primary content to Markdown.","{\n  \"mcpServers\": {\n    \"fetch\": {\n      \"command\": \"uvx\",\n      \"args\": [\"--with\", \"mcp\u003C2\", \"mcp-server-fetch\"]\n    }\n  }\n}\n",[1613,1614],"Does not execute client-side JavaScript; dynamic Single Page Applications will return empty templates.","Strict robots.txt enforcement will reject requests if autonomous crawling is disallowed by the target host.","When standard fetch fails on JS-rendered or bot-protected sites, Webfuse provides dynamic browser rendering and bypass.",[109,110,111],{},"The top 10 Model Context Protocol (MCP) servers for web scraping, crawling, SERP search, and markdown extraction, with tool schemas and configs for Claude Desktop, Cursor, and OpenCode.","Top 10 Web Scraping & Content Extraction MCP Servers for AI Agents (2026) | Webfuse","/mcp/thumbnails/web-scraping.png",[1622,1624,1626],{"title":430,"description":1623},"Evaluated strictly for implementation of official @modelcontextprotocol/sdk or vendor-backed Python/Node MCP libraries with standard tool schemas.",{"title":433,"description":1625},"Every callable tool name, input parameter, CLI launch string, and host config is verified directly against repository source code.",{"title":436,"description":1627},"Tested for output cleanliness: filtering boilerplate HTML and converting complex pages to token-efficient Markdown before agent consumption.","web-scraping","mcp/web-scraping","Blv2nXRjKgLh8T4FOZi09cWQaS-L7GPnDpzJwC17rJ8",1788261831578]