Лучшие MCP server для DevOps & Infrastructure в Codex

Enforces policy as code to orchestrate, schedule, and audit AI agent workloads with deterministic execution and verifiable lineage.

[mcp_servers.bernstein]
command = "uvx"
args = ["bernstein"]

Enforces policy-as-code governance for AI agents through deterministic scheduling and verifiable audit receipts.

[mcp_servers.bernstein]
command = "uvx"
args = ["bernstein"]

Enables AI assistants to interact with real-time infrastructure monitoring data, including metrics, logs, alerts, and anomaly detection.

[mcp_servers.mcp-server]
url = "https://app.netdata.cloud/api/v1/mcp"

Connects AI clients to GitHub so they can read repositories, manage issues and pull requests, and inspect workflows.

[mcp_servers.github-mcp-server]
command = "docker"
args = ["run", "-i", "--rm", "ghcr.io/github/github-mcp-server:2.1.0"]

Lets an AI client build, test, and manage Xcode projects and iOS simulators from the command line or MCP tools.

[mcp_servers.xcodebuildmcp]
command = "npx"
args = ["-y", "xcodebuildmcp"]

Enables AI clients to manage Kubernetes and OpenShift resources through direct API interaction, including pod operations, Helm charts, and Tekton pipelines.

[mcp_servers.kubernetes-mcp-server]
command = "npx"
args = ["-y", "kubernetes-mcp-server"]

Integrates with Terraform Registry and HCP Terraform APIs to automate Infrastructure as Code workflows and workspace management.

[mcp_servers.terraform-mcp-server]
command = "docker"
args = ["run", "-i", "--rm", "docker.io/hashicorp/terraform-mcp-server:1.4.1"]

Lets an AI client query and control an Unraid server's array, VMs, Docker containers and notifications via its GraphQL API.

[mcp_servers.unraid-mcp]
command = "uvx"
args = ["unraid-mcp"]

Lets AI tools manage files, websites, and serverless workers on a self-hostable open-source cloud computer.

[mcp_servers.puter]
url = "https://mcp.puter.com/"

Allows an AI client to manage UniFi Network devices, clients, firewall rules, VLANs, and VPNs through 210 specialized tools.

[mcp_servers.unifi-network-mcp]
command = "uvx"
args = ["unifi-network-mcp"]

Start, prompt, and read AI coding agent sessions across Claude Code, Codex, Gemini CLI, and other CLIs from one dashboard.

[mcp_servers.codeman]
command = "npx"
args = ["-y", "aicodeman"]

Create, execute commands in, and manage isolated Firecracker microVM sandboxes from any MCP client.

[mcp_servers.mcp]
command = "npx"
args = ["-y", "@superserve/mcp"]

Exposes all Dokploy API endpoints as tools to allow AI clients to manage projects, applications, databases, and Docker configurations.

[mcp_servers.mcp]
command = "npx"
args = ["-y", "@ahdev/dokploy-mcp"]

Lets an AI client manage GitLab projects, merge requests, issues, pipelines, wikis and releases through the GitLab API.

[mcp_servers.gitlab-mcp]
command = "npx"
args = ["-y", "@zereight/mcp-gitlab"]

Operate Proxmox VE infrastructure including VMs, LXCs, snapshots, and backups through MCP clients and AI agents.

[mcp_servers.proxmoxmcp-plus]
command = "uvx"
args = ["proxmox-mcp-plus"]

Automatically generates API tests and data mocks by recording real user traffic, database queries, and streaming events.

[mcp_servers.keploy]
url = "https://api.keploy.io/client/v1/mcp"

Lets AI agents control iOS and Android apps on simulators, emulators, and real devices through taps, typing, and accessibility snapshots.

[mcp_servers.mobile-mcp]
command = "npx"
args = ["-y", "@mobilenext/mobile-mcp"]