Explore
MCP servers and skills: search, filter by where they run, and compare.
MCP servers and skills: search, filter by where they run, and compare.
AWS Labs MCP server for Amazon EKS clusters, run locally over stdio.
OfficialPythonApache-2.0Needs AWS
AWS Labs MCP server for querying Amazon Aurora PostgreSQL databases, run locally over stdio.
OfficialPythonApache-2.0Needs AWS Aurora PostgreSQL
AWS Labs MCP server for infrastructure-as-code work on AWS, run locally over stdio.
OfficialPythonApache-2.0Needs AWS
Microsoft's Azure MCP Server covering 40+ Azure services in one server, installed from npm as @azure/mcp.
OfficialNode.jsMITNeeds Microsoft Azure
Web, news, image, video and local search through the Brave Search API.
OfficialNode.jsMITNeeds Brave Search API
Lets an agent control Chrome and inspect performance traces, network requests and console output via DevTools.
OfficialNode.jsApache-2.0No key
ClickHouse's official MCP server for listing databases and tables and running read queries on ClickHouse.
OfficialPythonApache-2.0Needs ClickHouse
Cloudflare's hosted MCP server that exposes its API through search and execute tools, plus docs search.
OfficialHosted by authorApache-2.0Needs Cloudflare
Fetches current, version-specific library documentation and code examples for prompts.
OfficialNode.jsMITNeeds Context7
Docker CLI plugin that runs MCP servers in isolated containers behind one gateway; powers Docker Desktop's MCP Toolkit.
OfficialSkillMITNeeds Docker
AWS Labs MCP server for DynamoDB tables and data, with a read-only option, run locally over stdio.
OfficialPythonApache-2.0Needs AWS DynamoDB
Reference test server that exposes sample prompts, resources and tools for checking MCP clients.
OfficialNode.jsMIT AND Apache-2.0No key
Neural web search and page content retrieval through Exa's hosted or local MCP server.
OfficialNode.jsMITNeeds Exa
Fetches a URL and converts the page content to markdown for LLM use.
OfficialPythonMIT AND Apache-2.0No key
Reads, writes and searches files inside directories the user explicitly allows.
OfficialNode.jsMIT AND Apache-2.0No key
Scrapes, maps, crawls and searches the web, returning clean markdown or structured data.
OfficialNode.jsMITNeeds Firecrawl
Reference server from the MCP project for reading diffs and history and making commits and branches in local Git repos.
OfficialPythonMIT AND Apache-2.0No key
GitHub's official MCP server for repos, issues, pull requests, Actions runs and code scanning alerts.
OfficialDockerMITNeeds GitHub
Grafana Labs' MCP server for dashboards, datasources, Prometheus and Loki queries, alerting, incidents and OnCall.
OfficialSkillApache-2.0Needs Grafana
Heroku's official MCP server for apps, dynos, add-ons, pipelines, logs and Heroku Postgres.
OfficialNode.jsApache-2.0Needs Heroku
Converts PDFs, Office files, HTML and other URIs to Markdown for LLM input.
OfficialPythonMITNo key
Persistent memory for agents as a local knowledge graph of entities, relations and observations.
OfficialNode.jsMIT AND Apache-2.0No key
MongoDB's MCP server for querying and indexing collections, with optional MongoDB Atlas admin tools.
OfficialNode.jsApache-2.0Needs MongoDB
MotherDuck's MCP server for querying local DuckDB files, MotherDuck cloud databases or S3 data.
OfficialPythonMITNeeds MotherDuck
Who stands behind it