# MCPfolio

> Search and compare Model Context Protocol servers by tools, workflows, and setup requirements. MCPfolio is for people and AI assistants comparing public MCP server listings, setup details, workflows, and source links.

MCPfolio publishes a searchable directory. Catalog details are labeled with their available sources and evidence; a listing does not mean MCPfolio has connected to that server.

## Core pages

- [MCP server directory](https://www.mcpfolio.com/index.md): Find MCP servers by product, category, tools, and workflow.
- [MCP server setup guides and comparisons](https://www.mcpfolio.com/articles.md): Source-backed guides to choosing and configuring MCP servers.
- [MCP server collections](https://www.mcpfolio.com/collections.md): Curated MCP server collections for common workflows.
- [MCP server workflows](https://www.mcpfolio.com/workflows.md): Research-led MCP workflows with steps, tools, and limitations.

## Guides

- [Best MCP Servers for Coding, Research, and Team Work](https://www.mcpfolio.com/articles/best-mcp-servers.md): Compare MCP servers for coding, web research, browser tasks, and team workflows. See selection criteria, documented setup, sources, and limitations for each pick.
- [YouTube Video Analysis MCPs: Transcripts, Frames, and OCR](https://www.mcpfolio.com/articles/youtube-video-analysis-mcp.md): Compare documented MCP approaches for YouTube dialogue and visuals, then follow a timestamped Codex workflow with clear limits on transcript, frame, and OCR evidence.

## Collections and workflows

- [Understand, change, and test a codebase](https://www.mcpfolio.com/collections/codebase-work.md): Cover repository work, issue planning, error triage, current library docs, and browser testing. Explain which tools can create or change data.
- [Research the web and inspect pages](https://www.mcpfolio.com/collections/web-research.md): Distinguish search and extraction from interactive browser automation; show local and hosted options clearly.
- [Connect an agent to application data](https://www.mcpfolio.com/collections/application-data.md): Compare database workflows and access requirements. Call out write, migration, and production-data risks.
- [Deploy and troubleshoot cloud apps](https://www.mcpfolio.com/collections/cloud-deploy-troubleshooting.md): Compare MCP servers for cloud deployments and troubleshooting. See which can inspect deployments, change configuration, or run deployment actions, plus required account access.
- [Coordinate team projects](https://www.mcpfolio.com/collections/team-projects.md): Coordinate work across team chat, shared documents, and issue tracking. Compare the workspace actions, permissions, and setup requirements each MCP server documents.
- [Move from design to implementation](https://www.mcpfolio.com/collections/design-to-implementation.md): Group design context, prototyping, site editing, diagrams, and asset discovery by role.
- [Prepare an evidence-backed support escalation](https://www.mcpfolio.com/workflows/support-escalation-evidence-packet.md): Join a customer report, error telemetry, and backlog context into a reviewed engineering handoff.
- [Audit Stripe subscriptions against app entitlements](https://www.mcpfolio.com/workflows/stripe-entitlement-discrepancy-audit.md): Compare current billing state with Supabase access state and return a reviewed exception report without changing either system.
- [Group duplicate customer requests without losing their sources](https://www.mcpfolio.com/workflows/deduplicate-customer-feature-requests.md): Turn scattered support requests into problem themes linked back to the original customer evidence.
- [Turn customer calls into reviewed notes and follow-ups](https://www.mcpfolio.com/workflows/customer-call-followup-packet.md): Carry decisions and promised next steps from a call transcript into the account record and task tracker.
- [Prioritize product problems with customer and usage evidence](https://www.mcpfolio.com/workflows/customer-feedback-priority-brief.md): Compare repeated customer pain with observed feature usage and backlog context before proposing roadmap priorities.
- [Draft an incident timeline from alerts and error evidence](https://www.mcpfolio.com/workflows/incident-timeline-postmortem-draft.md): Bring incident records and telemetry into a source-linked postmortem draft with uncertainty preserved.
- [Find postmortem actions that slipped past the incident](https://www.mcpfolio.com/workflows/postmortem-action-followthrough.md): Compare postmortem commitments with the issue tracker and code changes, then prepare a reviewed overdue-action report.
- [Find the runbook for an active incident and flag stale steps](https://www.mcpfolio.com/workflows/retrieve-and-check-incident-runbook.md): Use the incident service and symptoms to find the matching procedure, then surface gaps for an engineer to verify.
- [Prepare an evidence-based at-risk customer review](https://www.mcpfolio.com/workflows/prepare-at-risk-customer-review.md): Bring together recent usage, support context, and billing identifiers for one customer before a renewal or health review.
- [Find broken product analytics events before trusting a funnel](https://www.mcpfolio.com/workflows/audit-analytics-events-against-tracking-contract.md): Compare an agreed tracking contract with analytics taxonomy, event data, and existing engineering work before making product decisions.

## MCP server listings

- [Atlassian Rovo MCP](https://www.mcpfolio.com/servers/atlassian-jira-confluence.md): A cloud-hosted MCP server that connects AI clients to Jira, Confluence, Jira Service Management, Bitbucket, Loom, and other Atlassian Cloud data through Teamwork Graph.
- [Context7 MCP](https://www.mcpfolio.com/servers/context7.md): Finds current, version-aware library documentation and code examples, then returns relevant context to an AI client.
- [Sequential Thinking MCP](https://www.mcpfolio.com/servers/sequential-thinking.md): A local MCP server that exposes a structured thinking tool for iterative problem solving, revisions, and branches in a reasoning process.
- [Playwright MCP](https://www.mcpfolio.com/servers/playwright-browser-automation.md): A local MCP server that lets AI clients automate browsers through Playwright and structured accessibility snapshots.
- [Terraform MCP](https://www.mcpfolio.com/servers/hashicorp-terraform.md): A local MCP server for Terraform Registry research and, when configured, HCP Terraform or Terraform Enterprise workspace operations.
- [Bright Data MCP](https://www.mcpfolio.com/servers/bright-data.md): Connects AI clients to Bright Data's web-data services for search, webpage extraction, scraper APIs, and browser automation.
- [Mem0 MCP](https://www.mcpfolio.com/servers/mem0.md): Connects AI clients to a hosted Mem0 account for saving, searching, retrieving, updating, and managing user or agent memories.
- [Tableau MCP](https://www.mcpfolio.com/servers/tableau-mcp.md): Connects AI agents to Tableau Cloud or Server to query curated data, inspect workbooks and views, analyze Pulse metrics, and manage Tableau content.
- [GitHub MCP](https://www.mcpfolio.com/servers/github.md): GitHub's official MCP server gives AI clients access to GitHub repositories, issues, pull requests, actions, users, and other capabilities through configurable toolsets.
- [Canva MCP](https://www.mcpfolio.com/servers/canva.md): Canva's remote MCP server lets authenticated AI assistants search, create, edit, export, and collaborate on Canva designs, with available actions limited by each user's Canva permissions and plan.
- [Chrome DevTools MCP](https://www.mcpfolio.com/servers/chrome-devtools.md): Google's Chrome DevTools MCP connects an AI coding agent to a live Chrome browser for browser automation, debugging, network inspection, performance analysis, and related DevTools workflows.
- [Desktop Commander MCP](https://www.mcpfolio.com/servers/desktop-commander.md): An independent local MCP server for filesystem operations, terminal sessions, process control, code editing, and structured document handling. The entry here is the local @wonderwhy-er/desktop-commander package; the separate Desktop Commander Remote service is not merged into this implementation record.
- [Knowledge Graph Memory](https://www.mcpfolio.com/servers/knowledge-graph-memory.md): An official local memory server that persists entities, relations, and observations as a knowledge graph in a JSONL file and exposes graph operations through MCP tools and a resource.
- [Weaviate MCP](https://www.mcpfolio.com/servers/weaviate.md): A built-in Weaviate MCP server that exposes collection configuration, tenant listing, hybrid vector/keyword search, and optionally object upserts through the database's REST API host.
- [Webflow MCP](https://www.mcpfolio.com/servers/webflow.md): Webflow's hosted MCP server connects AI clients to Webflow sites and workspaces for managing site data, CMS, assets, design systems, analytics, and selected Designer actions.
- [Firecrawl MCP](https://www.mcpfolio.com/servers/firecrawl.md): Firecrawl's MCP server gives AI clients tools to search the web, scrape and parse content, map or crawl sites, interact with pages, and run research or monitoring workflows.
- [Retool MCP](https://www.mcpfolio.com/servers/retool.md): Retool's hosted MCP server lets authorized AI clients build and inspect Retool apps, work with organization users and settings, and manage connected resources.
- [Browserless MCP](https://www.mcpfolio.com/servers/browserless.md): Browserless provides a hosted MCP server for browser automation, scraping, search, crawling, page interaction, exports, audits, and account usage.
- [Gemini Cloud Assist MCP](https://www.mcpfolio.com/servers/gemini-cloud-assist.md): Google Cloud's agentic MCP service provides assistance for cloud troubleshooting, infrastructure design, cost optimization, and selected cloud operations.
- [Filesystem MCP](https://www.mcpfolio.com/servers/filesystem.md): A local MCP reference server that provides controlled file and directory operations within explicitly allowed directories.
- [Jira MCP](https://www.mcpfolio.com/servers/jira.md): The discovery entry appears to refer to Atlassian's hosted Rovo MCP server, which exposes Jira tools as part of the broader Atlassian Cloud integration.
- [Miro MCP](https://www.mcpfolio.com/servers/miro.md): Miro's hosted MCP server lets authorized AI clients read, search, summarize, create and update Miro boards, including diagrams, documents, tables, code widgets and comments.
- [Notion MCP](https://www.mcpfolio.com/servers/notion.md): Notion's hosted MCP server connects AI clients to the authorized workspace for search, reading, page/database creation and updates, file handling, and other workspace workflows.
- [Ramp MCP](https://www.mcpfolio.com/servers/ramp.md): Ramp hosted MCP/CLI exposes finance and spend-management tools under the authenticated Ramp identity, with runtime availability governed by permissions, scopes, admin controls and environment.
- [21st MCP (formerly Magic MCP)](https://www.mcpfolio.com/servers/magic.md): The listing's Magic MCP has been replaced by 21st MCP, an authenticated design/code component server for searching components and logos, retrieving components, inspiration and account-dependent UI generation.
- [Figma MCP](https://www.mcpfolio.com/servers/figma.md): Figma's MCP server gives supported AI clients design context from Figma files and can create or edit native Figma content, generate diagrams, and transfer live UI into Figma.
- [Typeform MCP](https://www.mcpfolio.com/servers/typeform.md): Typeform's hosted MCP server lets AI clients create and edit forms, build automations, manage contacts, and query response insights through Typeform OAuth.
- [Brave Search MCP](https://www.mcpfolio.com/servers/brave-search.md): Brave's official local MCP server connects AI clients to Brave Search API capabilities including web, local business, place, image, video, news, LLM-context, and AI-answer search.
- [Magic UI MCP](https://www.mcpfolio.com/servers/magic-ui.md): Magic UI's official MCP server searches and retrieves components from the live Magic UI registry, including installation instructions, source, related items, and examples.
- [DeepWiki MCP](https://www.mcpfolio.com/servers/deepwiki.md): DeepWiki's hosted MCP server retrieves documentation generated for public GitHub repositories and answers repository questions grounded in that documentation.
- [Supabase MCP](https://www.mcpfolio.com/servers/supabase-mcp-server.md): Supabase's hosted MCP server connects AI coding clients to Supabase projects for database, documentation, debugging, development, Edge Functions, Storage, branching, and account operations.
- [Azure DevOps MCP](https://www.mcpfolio.com/servers/azure-devops.md): Microsoft's Azure DevOps MCP server gives compatible AI clients access to organization projects, work items, repositories, pull requests, pipelines, test plans, and wikis through hosted or local server modes.
- [Microsoft Learn MCP](https://www.mcpfolio.com/servers/microsoft-docs.md): Microsoft Learn's hosted MCP server lets AI clients search Microsoft technical documentation, fetch full documentation articles, and search Microsoft/Azure code samples.
- [Jotform MCP](https://www.mcpfolio.com/servers/jotform.md): Jotform's hosted MCP server lets AI clients find forms and apps, create or edit forms, retrieve form/submission data, assign forms, and analyze submission trends.
- [Browser Use MCP](https://www.mcpfolio.com/servers/browser-use.md): Browser Use exposes browser automation to AI clients through either a local Python MCP server or a hosted cloud MCP service, with tools for browser actions and autonomous tasks.
- [Confluence MCP](https://www.mcpfolio.com/servers/confluence.md): The directory's Confluence entry appears to refer to Atlassian's hosted Rovo MCP server, which provides access to Confluence alongside Jira and other connected Atlassian Cloud products.
- [Exa MCP](https://www.mcpfolio.com/servers/exa-search.md): Exa's hosted MCP server provides web search and page fetching for AI clients, with optional advanced search and multi-step Exa Agent research tools.
- [Icons8 MCP](https://www.mcpfolio.com/servers/icons8.md): A hosted MCP server for searching and retrieving Icons8 icons and animations for design and coding workflows.
- [Linear MCP](https://www.mcpfolio.com/servers/linear.md): Linear's hosted MCP server connects AI clients to Linear data and workflows for finding, creating, and updating issues, projects, and comments.
- [Parallel Search MCP](https://www.mcpfolio.com/servers/parallel-search.md): Parallel's hosted MCP adds web search and URL content extraction to AI clients, with anonymous Fast mode and account-backed settings for higher limits.
- [Perplexity API MCP](https://www.mcpfolio.com/servers/perplexity.md): Perplexity's official MCP server gives AI clients real-time web search, conversational search, deep research, and reasoning through hosted or local connections.
- [Ahrefs MCP](https://www.mcpfolio.com/servers/ahrefs.md): Ahrefs hosted MCP gives supported AI clients scoped access to Ahrefs marketing and search data, with selectable tool groups.
- [Firebase MCP](https://www.mcpfolio.com/servers/firebase.md): Google's local Firebase CLI MCP server lets AI development clients inspect and manage Firebase projects and app configuration, with additional product-specific tools, prompts, and resources.
- [Sentry MCP](https://www.mcpfolio.com/servers/sentry.md): Sentry's hosted MCP server connects AI clients to Sentry organizations and projects to search errors, inspect performance, triage issues, access documentation, and manage project context.
- [Hugging Face MCP](https://www.mcpfolio.com/servers/huggingface.md): Hugging Face's hosted MCP connects AI clients to Hub models, datasets, Spaces, papers, documentation search, and optionally selected community Gradio Spaces.
- [Semrush MCP](https://www.mcpfolio.com/servers/semrush.md): Semrush's hosted MCP gives compatible AI clients access to Semrush SEO, website traffic, market, and project data for keyword research and competitive analysis.
- [Grep](https://www.mcpfolio.com/servers/grep.md): Hosted MCP endpoint for searching indexed public GitHub repositories and returning matching code snippets, with query filters such as language, repository, and file path.
- [MasterGo Design](https://www.mcpfolio.com/servers/mastergo-design.md): Connects AI clients to MasterGo design files to retrieve design DSL and metadata, inspect page layers, extract SVG assets, and support design/code workflows.
- [Netlify MCP](https://www.mcpfolio.com/servers/netlify.md): Netlify's official MCP connects AI agents to Netlify platform expertise and CLI-backed workflows, with remote hosted and local package options.
- [Tavily MCP](https://www.mcpfolio.com/servers/tavily-search.md): Tavily's MCP server gives AI agents real-time web search, URL extraction, site mapping, and crawling through remote or local connections.

The list above includes up to 50 combined collection and workflow entries and 50 server listings. Use the Markdown catalog for all addressable public pages.

## Machine-readable resources

- [Full content](https://www.mcpfolio.com/llms-full.txt): bounded collection of public page Markdown.
- [Markdown catalog](https://www.mcpfolio.com/api/md/_catalog): registry of public HTML and Markdown URLs.
- [Sitemap](https://www.mcpfolio.com/sitemap.xml)
- [AI usage policy](https://www.mcpfolio.com/ai.txt)
- [MCP setup guide](https://www.mcpfolio.com/mcp.md)
- [MCPfolio setup flow](https://www.mcpfolio.com/mcpfolio-setup.md)
- [MCP server card](https://www.mcpfolio.com/.well-known/mcp/server-card.json)

Keep this file in sync with /ai.txt and /robots.txt.