Use this skill whenever the user needs to manage VMware NSX networking — segments, gateways, NAT, routing, and IP pools. Directly handles: create/manage network segments, configure Tier-0/Tier-1 gateways, set up NAT rules, manage static routes, configure IP pools, check transport node and edge cluster health. Always use this skill for "create segment", "set up gateway", "create NAT rule", "check network health", "troubleshoot connectivity", or any NSX/networking/segment task. Do NOT use for DFW firewall rules or security groups (use vmware-nsx-security), VM lifecycle (use vmware-aiops), or AVI/ALB load balancing (use vmware-avi). For multi-step workflows use vmware-pilot.
Use this skill whenever the user needs to manage VMware NSX networking — segments, gateways, NAT, routing, and IP pools. Directly handles: create/manage network segments, configure Tier-0/Tier-1 gateways, set up NAT rules, manage static routes, configure IP pools, check transport node and edge cluster health. Always use this skill for "create segment", "set up gateway", "create NAT rule", "check network health", "troubleshoot connectivity", or any NSX/networking/segment task. Do NOT use for DFW firewall rules or security groups (use vmware-nsx-security), VM lifecycle (use vmware-aiops), or AVI/ALB load balancing (use vmware-avi). For multi-step workflows use vmware-pilot.
vmware-policy auto-installed as Python dependency (provides @vmware_tool decorator and audit logging). All write operations audited to ~/.vmware/audit.db. Credentials: Each NSX Manager target requires a per-target password env var in ~/.vmware-nsx/.env following the pattern VMWARE_NSX_<TARGET_NAME_UPPER>_PASSWORD. Also supports certificate-based auth. Passwords are never logged or echoed. Destructive operations: Segment/gateway/NAT delete require double confirmation + --dry-run. Segment delete checks for connected ports, gateway delete checks for connected segments. VMWARE_AUDIT_APPROVED_BY is an optional attestation recorded in the audit row; it is not a gate and does not carry credentials. No webhooks, no outbound network calls, no guest operations. Local only: stdio MCP + NSX Policy API (HTTPS 443). SSL bypass: verify_ssl is on by default; false option for self-signed certs in lab environments only. Transitive dependencies: Only vmware-policy (audit/policy). No post-install scripts or background services.
VMware NSX
Disclaimer: This is a community-maintained open-source project and is not affiliated with, endorsed by, or sponsored by VMware, Inc. or Broadcom Inc. "VMware" and "NSX" are trademarks of Broadcom. Source code is publicly auditable at github.com/zw008/VMware-NSX under the MIT license.
VMware NSX networking management — 33 MCP tools for segments, gateways, NAT, routing, and IPAM.
Create an App Network (Segment + T1 Gateway + NAT)
Pre-flight (judgment, not blind sequence):
Subnet conflict check: scan inventory list-segments and networking list-ip-pools for any overlap with the proposed CIDR. Overlapping subnets cause asymmetric routing or silent blackholing — NSX will not warn you.
Edge cluster capacity: confirm chosen --edge-cluster is healthy (inventory list-edge-clusters + health edge-cluster-status <id>) and not at SR (Service Router) limit. A fully-loaded edge cluster will accept the T1 creation but routing will fail.
T0 uplink: the parent T0 must already be configured with BGP/static routes upstream — otherwise SNAT works internally but external traffic goes nowhere.
NAT IP: --translated IP must be from a routable address pool announced by T0; using a random IP creates a half-working network.
Always --dry-run first — once a segment is attached to running VMs, deleting it requires detaching every port.
Steps:
vmware-nsx gateway create-tier1 app-t1 --name app-t1 --edge-cluster <ec-path> --tier0 <t0-path> --dry-run → review, then run for real
Verify end-to-end: inventory list-segments, networking list-nat-rules app-t1, AND test with a VM attached to the new segment
On failure: a connection error or HTTP error prints a single teaching line (e.g. 403 → check NSX role privileges; 404 → run the matching list command for the exact ID). Run vmware-nsx doctor to verify connectivity and credentials, fix, and re-run the failed step — earlier completed steps are idempotent PUTs and safe to re-apply.
Check Network Health
Judgment: don't just enumerate health endpoints — correlate them. The order below maps cause to symptom: if manager is down, transport nodes will look down too (false positive); fix top-down.
vmware-nsx health manager-status — if any manager node is DEGRADED or DOWN, stop here and resolve before trusting downstream signals
vmware-nsx inventory list-transport-nodes then health transport-node-status <id> for any node not UP — flag nodes down ≥ 5 min; transient blips are normal
vmware-nsx inventory list-edge-clusters then health edge-cluster-status <id> — verify SR placement is balanced; one edge holding 80% of SRs is a single point of failure
vmware-nsx health alarms --severity HIGH (repeat with CRITICAL) — severity filter is exact-match, not "and above"
Cross-check with vmware-monitor for vSphere host events — a host losing connection to vCenter often masquerades as an NSX problem
Troubleshoot VM Connectivity
Judgment: connectivity failures happen at one of three layers. Identify which layer first, then drill — don't probe randomly.
Layer 1 — VM-to-segment: VM has no segment, wrong vNIC, or port admin-down → troubleshoot vm-segment + troubleshoot port-status
Layer 2 — segment-to-gateway: segment not attached to T1, T1 not connected to T0 → inventory get-tier1 shows no Tier-0 path
Layer 3 — gateway-to-upstream: T0 BGP/static missing or SNAT not configured → networking bgp-neighbors, networking list-nat-rules
Steps (stop as soon as the failing layer is identified):
Layer 1: troubleshoot vm-segment my-vm-01 → if no port, check vSphere vNIC binding first
Layer 1: troubleshoot port-status <segment-id> → admin-down or DFW-blocked? If DFW, jump to vmware-nsx-security
Layer 2: inventory get-tier1 app-t1 → Tier-0 path present and route advertisement enabled? If not, T1↔T0 link broken
Layer 3: networking list-nat-rules app-t1 → SNAT rule covers the source CIDR? Mis-typed CIDR is the most common cause
Multi-Target Operations
All commands accept --target <name> to operate against a specific NSX Manager from your config (default: the first target in config.yaml), e.g. vmware-nsx inventory list-segments --target nsx-prod.
Usage Mode
Scenario
Recommended
Why
Local/small models (Ollama, Qwen)
CLI
~2K tokens vs ~8K for MCP
Cloud models (Claude, GPT-4o)
Either
MCP gives structured JSON I/O
Automated pipelines
MCP
Type-safe parameters, structured output
MCP Tools (33 — 20 read, 13 write)
All MCP tools accept an optional target parameter to select which NSX Manager to connect to.
Category
Tool
Type
Description
Segment
list_segments
Read
List all segments with type, subnet, admin state, port count
get_segment
Read
Get segment details including ports and subnet config
create_segment
Write
Create overlay or VLAN segment with subnet and gateway
Delete a Tier-1 gateway (removes default locale-service first)
NAT
list_nat_rules
Read
List NAT rules on a Tier-1 gateway
create_nat_rule
Write
Create SNAT/DNAT/reflexive NAT rule on a gateway
delete_nat_rule
Write
Delete a NAT rule
Static Routes
list_static_routes
Read
List static routes on a Tier-1 gateway
create_static_route
Write
Add a static route with network and next-hop
delete_static_route
Write
Remove a static route
IP Pools
list_ip_pools
Read
List IP pools with usage summary
get_ip_pool_usage
Read
Show allocation usage for a pool
create_ip_pool
Write
Create a new IP address pool with allocation ranges
delete_ip_pool
Write
Permanently delete an IP address pool
Fabric
list_transport_zones
Read
List transport zones with type (OVERLAY/VLAN)
list_transport_nodes
Read
List transport nodes with node type and status
list_edge_clusters
Read
List edge clusters with member count and deployment type
Health
list_nsx_alarms
Read
List active NSX alarms filtered by severity
get_transport_node_status
Read
Transport node connectivity and config status
get_edge_cluster_status
Read
Edge cluster member status and failover config
get_nsx_manager_status
Read
NSX Manager cluster health and node roles
Troubleshoot
get_logical_port_status
Read
Realized state of all ports on a segment
get_segment_port_for_vm
Read
Find which segment a VM is connected to by display name
Write tools require explicit parameters and are audit-logged. Dry-run preview (--dry-run) is a CLI feature; MCP write tools execute directly.
List results are envelopes — read truncated before you summarise
Every list-returning tool above returns {items, returned, limit, total, truncated, hint}, not a bare array. Rows live under items: empty items with truncated: false means the query genuinely matched nothing — report that, not a tool failure. truncated: true means more rows exist — never describe the result as the complete set; re-query as hint instructs. Field semantics, total sourcing, and an example payload: references/capabilities.md.
Local & Small Models
Running with local or small models? See references/agent-guardrails.md for explicit operating rules that keep tool calls reliable.
Full CLI reference with all options and output formats: see references/cli-reference.md
Troubleshooting
"Segment not found" when querying
Segment display names and Policy API IDs can differ. Use vmware-nsx inventory list-segments to get the exact ID. The Policy API uses the segment id field, not display_name. Common mistakes: using the display name with spaces instead of the hyphenated ID.
NAT rule creation fails with "gateway not found"
NAT rules are created on Tier-1 gateways (or Tier-0 for some topologies). Verify the gateway name with vmware-nsx inventory list-tier1s. The gateway must have an edge cluster assigned for NAT to function.
BGP neighbor shows "Connect" or "Active" state
The BGP session is not established. Common causes:
Peer IP unreachable from the edge node — check physical uplinks and VLAN config
ASN mismatch — compare local and remote ASN in bgp-neighbors output
MD5 password mismatch — verify authentication settings on both sides
Transport node status "degraded"
A transport node in degraded state has partial connectivity. Steps:
Check vmware-nsx health transport-nodes for the specific failure reason
Common cause: tunnel endpoint (TEP) unreachable — verify underlay MTU (minimum 1600 for Geneve)
Check NTP sync between NSX Manager and transport nodes
If recently upgraded, verify the host switch config matches NSX Manager expectations
"Password not found" error
The password environment variable is missing. Variable names follow the pattern VMWARE_NSX_<TARGET_NAME_UPPER>_PASSWORD where hyphens become underscores. Example: target nsx-prod needs VMWARE_NSX_NSX_PROD_PASSWORD. Check your ~/.vmware-nsx/.env file.
Safety
Read-heavy: 20 of 33 tools are read-only (list, get, status, health, troubleshoot)
Audit logging: All operations logged to ~/.vmware/audit.db (SQLite WAL, via vmware-policy) with timestamp, user, target, operation, parameters, and result
Double confirmation: CLI write commands require two separate confirmation prompts before executing
Dry-run mode: All CLI write commands support --dry-run to preview API calls without executing (MCP write tools execute directly and are audit-logged)
Dependency checks: Segment delete checks for connected ports; gateway delete checks for connected segments; prevents accidental cascade failures
Input validation: CIDR networks validated, IP addresses checked, gateway existence verified before NAT/route operations
Prompt injection defense: NSX object names returned from the API are sanitized via _sanitize() — strips control characters, truncates to 500 chars
Credential safety: Passwords loaded only from environment variables (.env file), never from config.yaml
No firewall operations: Cannot create, modify, or delete DFW rules, security groups, or IDS/IPS policies — that scope belongs to vmware-nsx-security
Setup
uv tool install vmware-nsx-mgmt
vmware-nsx init # writes ~/.vmware-nsx/config.yaml + .env (chmod 600), then verifies
vmware-nsx doctor
All tools are automatically audited via vmware-policy. Audit logs: vmware-audit log --last 20
Full setup guide with multi-target config, MCP server setup, and Docker: see references/setup-guide.md
Architecture
User (natural language)
|
AI Agent (Claude Code / Goose / Cursor)
| reads SKILL.md
vmware-nsx CLI or MCP server (stdio transport)
| NSX Policy API (REST/JSON over HTTPS)
NSX Manager
|
Segments / Gateways / NAT / Routes / IP Pools / Transport Nodes
Audit & Safety
All operations are automatically audited via vmware-policy (@vmware_tool decorator):
Every tool call logged to ~/.vmware/audit.db (SQLite, framework-agnostic)
