ClickHouse's observability MCP server — investigate logs, traces and metrics with semantic tools rather than raw SQL, hosted or self-hosted.
Nutanix V4 API MCP Server
Summary
Nutanix's official open-source MCP server: 1,000+ Prism Central V4 API operations across 19 namespaces, read-only by default and audit-logged.
Features
- 1,000+ Prism Central V4 API operations across 19 namespaces
- One `<namespace>_execute` tool per namespace instead of per-endpoint tool sprawl
- Schemas resolved from your own Prism Central version, not a generic spec
- Read-only by default, with every call audit-logged
- Inherits Prism V4 API Gateway RBAC and rate limiting
- Air-gapped deployment supported; stdio transport, Python 3.11+
Installation
Set up this MCP server in your favorite AI agent — copy a ready-made configuration below.
Description
Nutanix's official Model Context Protocol server exposes the Prism Central V4 REST APIs — Nutanix's modern management plane for multi-cluster infrastructure — as tools an AI assistant can call in natural language. Announced on 10 August 2026, it lets Claude, Cursor, GitHub Copilot and any other MCP client run hybrid-cloud operations without a human hand-translating intent into API payloads.
The architecture is the interesting part. Rather than shipping one tool per endpoint and flooding the model's context, the server registers a single <namespace>_execute tool per API namespace — 19 of them, covering vmm (VM lifecycle), clustermgmt, networking (AHV, BGP, vSwitch, VPC, subnets), storage, volumes, iam, microseg (network security policy), dataprotection, datapolicies, lifecycle (firmware and software upgrades), monitoring, security, files, objects, licensing, prism, aiops, opsmgmt and multidomain. Together they reach over 1,000 V4 operations.
Schemas are resolved from your own cluster, not from a generic latest spec: nutanix-mcp init queries the connected Prism Central, discovers which namespaces that PC version actually exposes, and registers only those tools. An agent asking about a capability your deployment does not have simply will not find it.
Governance is the reason this exists in vendor-supported form. All traffic routes through the existing Prism V4 API Gateway, so agent activity inherits the same role-based access control, audit logging and rate limiting that already govern human and scripted access. The server is read-only by default, every call is audit-logged, and air-gapped deployments are supported.
It runs over stdio transport and requires Python 3.11 or later. Version 0.8 is a Technical Preview and Nutanix explicitly does not support it for production use — treat it as an evaluation tool for now. Apache-2.0 licensed.
Related MCP Servers
Vanta's official remote MCP server — query controls, tests, vendors, vulnerabilities and policies across SOC 2 and ISO 27001, and remediate failing tests from your agent.
Turso's hosted MCP server — manage Turso Cloud databases, branch them point-in-time and run SQL from a coding agent, authorised over OAuth with no API token to paste.
Google's official Firebase MCP server — 70+ tools across Firestore, Auth, Crashlytics, Cloud Functions, Storage, Messaging and Remote Config, shipped inside the Firebase CLI.
