MCP Servers Directory
Comprehensive guide to Model Context Protocol servers across cloud providers
Connect to MCP ServersWhat is Model Context Protocol (MCP)?
Model Context Protocol (MCP) is an open standard that enables AI agents and Large Language Models (LLMs) to securely connect with external tools, APIs, and data sources—across cloud, on-premise, and SaaS environments. MCP acts as a universal adapter, letting AI systems access live business data, trigger workflows, and perform actions, all through a single, standardized protocol.
Read the official MCP specification |
Comprehensive MCP guide (Medium) |
Quick read: What is MCP?
Why Use MCP?
- Unified Integration: Replace dozens of custom APIs with one protocol for all your AI tools and data sources.
- Security & Control: Explicit permissions, local deployment, and full auditability. You decide what your AI can access. MCP Security Best Practices
- Scalability: Add new tools or data sources instantly—no need to rewire your AI stack.
- Future-Proof: Adopted by leading AI platforms (Anthropic, OpenAI, Google, Microsoft) and a fast-growing open-source ecosystem.
How Does MCP Work?
- MCP Clients (AI apps, IDEs, chatbots) connect to MCP Servers (tool/data adapters) using a secure, standardized protocol.
- Servers expose tools, data, and workflows. Clients discover and invoke these capabilities dynamically—no hardcoding required.
- Supports local, cloud, and hybrid deployments. Getting started with MCP (Medium)
Security & Best Practices
- Follow the principle of least privilege: only expose what’s needed.
- Use strong authentication, encrypted transport, and regular audits. MCP Security 101 (Protect AI)
- Vet all MCP servers before connecting—prefer official or vendor-maintained integrations.
- For enterprise deployments, see Securing MCP in Enterprise AI
How to Use This Directory
- Browse the tables below to find MCP servers and toolboxes by provider and category.
- Each entry includes a description and direct links to documentation, GitHub, or NPM.
- For a deep dive on MCP Toolbox for Databases, see the official documentation and GitHub repository.
AWS MCP Servers
| Server Name | Type | Deployment | Description | Links |
|---|---|---|---|---|
| S3 MCP | Storage | Self-hosted | Access and manage AWS S3 buckets and objects. | awslabs/mcp (S3) |
| DynamoDB MCP | Database | Self-hosted | Query and manage DynamoDB tables. | awslabs/mcp (DynamoDB) |
| ECS MCP | Container | Self-hosted | Deploy and manage AWS ECS containers. | awslabs/mcp (ECS) |
| EKS MCP | Kubernetes | Self-hosted | Manage AWS EKS Kubernetes clusters. | awslabs/mcp (EKS) |
| Lambda MCP | Serverless | Self-hosted | Deploy and manage AWS Lambda functions. | awslabs/mcp (Lambda) |
| Bedrock MCP | AI/ML | Self-hosted | Access AWS Bedrock foundation models. | awslabs/mcp (Bedrock) |
| RDS/Aurora MCP | Database | Self-hosted | Manage Aurora MySQL/Postgres and RDS Data API. | awslabs/mcp (Aurora MySQL) |
| Serverless MCP | Serverless | Self-hosted | Manage AWS SAM, Lambda, API Gateway, and more. | awslabs/mcp (Serverless) |
| ECR MCP | Container Registry | Self-hosted | Manage AWS Elastic Container Registry. | awslabs/mcp (ECR) |
| CloudFormation MCP | IaC | Self-hosted | Manage AWS CloudFormation stacks. | awslabs/mcp (CloudFormation) |
| CDK MCP | IaC | Self-hosted | Manage AWS CDK workflows. | awslabs/mcp (CDK) |
Azure MCP Servers
| Server Name | Type | Deployment | Description | Links |
|---|---|---|---|---|
| Azure Blob Storage MCP | Storage | Self-hosted | Access and manage Azure Blob Storage containers. | NPM |
| Azure Functions MCP | Serverless | Self-hosted | Deploy and manage Azure Functions. | NPM |
| Azure OpenAI MCP | AI/ML | Self-hosted | Access Azure OpenAI services. | NPM |
| Cosmos DB MCP | Database | Self-hosted | Query and manage Cosmos DB collections. | NPM |
APM MCP Servers
| Server Name | Type | Deployment | Description | Links |
|---|---|---|---|---|
| Dynatrace MCP | APM | Self-hosted | Monitor and manage Dynatrace APM resources via MCP. | GitHub |
Infrastructure as Code (IaC) MCP Servers
| Server Name | Type | Deployment | Description | Links |
|---|---|---|---|---|
| Terraform MCP | IaC | Self-hosted | Manage Terraform infrastructure as code. | NPM |
| Ansible MCP | IaC/Automation | Self-hosted | Execute Ansible playbooks and tasks. | NPM |
Source Control Management (SCM) MCP Servers
Self-Hosted MCP Servers
| Server Name | Type | Deployment | Description | Links |
|---|---|---|---|---|
| Filesystem MCP | File System | Self-hosted | Access local file system with security controls. | NPM |
| Postgres MCP | Database | Self-hosted | Query and manage PostgreSQL databases. | NPM |
| Memory MCP | Memory | Self-hosted | Persistent memory storage for conversations. | NPM |
| Docker MCP | Container | Self-hosted | Manage Docker containers and images. | NPM |
| Kubernetes MCP | Orchestration | Self-hosted | Manage Kubernetes clusters and resources. | NPM |
Google Cloud MCP Servers
| Server Name | Type | Deployment | Description | Links |
|---|---|---|---|---|
| Cloud Spanner MCP | Database | Self-hosted | No official MCP server available. See docs to build your own. | Spanner Docs | MCP Hosting Guide |
| CloudSQL (Postgres) MCP | Database | Self-hosted | No official MCP server. Use PostgreSQL MCP server for CloudSQL Postgres. | PostgreSQL MCP | CloudSQL Docs |
| BigQuery MCP | Data Analytics | Self-hosted | No official MCP server available. See docs to build your own. | BigQuery Docs | MCP Hosting Guide |