Kubernetes MCP
containersTalks to the Kubernetes API directly instead of shelling out to kubectl, so an agent can list contexts, read pod logs, and apply manifests without parsing command output
Kubernetes MCP vs Kubernetes MCP (kubectl)
Talks to the Kubernetes API directly instead of shelling out to kubectl, so an agent can list contexts, read pod logs, and apply manifests without parsing command output On the other side: wraps kubectl so an agent can list resources, read logs, and inspect events
Installnpx -y kubernetes-mcp-server@latestnpx -y mcp-server-kubernetes
SourceOfficialCommunity
Deciding factorYou want predictable behavior across clusters and a smaller blast radius.versus: Your workflow already depends on kubectl flags and custom contexts.
Bottom line
Direct API unless your workflow is already shaped like kubectl.
The short answer
Both drive a cluster. The direct-API server talks to the Kubernetes API without shelling out, so it does not depend on your kubeconfig having the right context active. The kubectl wrapper inherits every context you have configured, which is convenient until it is not.
Facts
| Option | Install | Source | Link |
|---|---|---|---|
| Kubernetes MCP | npx -y kubernetes-mcp-server@latest | Official | Repository ↗ |
| Kubernetes MCP (kubectl) | npx -y mcp-server-kubernetes | Community | Repository ↗ |
Pick Kubernetes MCP if
You want predictable behavior across clusters and a smaller blast radius.
Pick Kubernetes MCP (kubectl) if
Your workflow already depends on kubectl flags and custom contexts.
What would change our mind
A project going archived, or a maintainer changing the install path, flips this comparison faster than any feature release. Both entries are re-checked weekly, and the date above tells you how much to trust the details.
Read the full entries
More head-to-heads
Filesystem vs Desktop Commander
Gives an agent scoped read, write, and search access to directories you explicitly allow
Comparison
Git vs GitHub
Read-only access to a repository's history: status, diffs, commit logs, and blame
Comparison
LaunchDarkly MCP vs Unleash MCP
Reads and changes feature flags through the LaunchDarkly API, so an agent can turn a flag off during an incident instead of asking a human to open a dashboard
Comparison
Sentry vs Datadog MCP
Pulls issues, stack traces, and event context into the agent so it can triage production errors with the actual payload in front of it
Comparison
Atlassian Jira MCP vs Linear MCP
Connects to Jira for issue lookup, JQL search, and comments
Comparison
Contentful MCP vs Notion MCP
Reads content types and entries and can create drafts in a Contentful space
Comparison