CloudStack MCP Server
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Server capabilities have not been inspected yet.
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| list_virtual_machinesC | List virtual machines in CloudStack |
| get_virtual_machineB | Get details of a specific virtual machine |
| start_virtual_machineC | Start a virtual machine |
| stop_virtual_machineC | Stop a virtual machine |
| reboot_virtual_machineC | Reboot a virtual machine |
| destroy_virtual_machineA | Destroy a virtual machine using proper workflow: stop → destroy → expunge. Handles VMs in any state including Error. (DESTRUCTIVE - cannot be undone) |
| deploy_virtual_machineC | Deploy a new virtual machine |
| scale_virtual_machineC | Scale virtual machine (change service offering) |
| migrate_virtual_machineC | Migrate virtual machine to another host |
| reset_password_virtual_machineC | Reset password for virtual machine |
| change_service_offering_virtual_machineC | Change service offering for virtual machine |
| list_virtual_machine_metricsC | List virtual machine performance metrics |
| list_volumesC | List storage volumes |
| create_volumeC | Create a new volume |
| attach_volumeC | Attach volume to virtual machine |
| detach_volumeC | Detach volume from virtual machine |
| resize_volumeC | Resize a volume |
| create_snapshotC | Create a snapshot of a volume |
| list_snapshotsC | List volume snapshots |
| list_networksC | List networks |
| create_networkC | Create a new network |
| list_public_ip_addressesC | List public IP addresses |
| associate_ip_addressC | Acquire a new public IP address |
| enable_static_natC | Enable static NAT for an IP to a VM |
| create_firewall_ruleC | Create a firewall rule |
| list_load_balancer_rulesC | List load balancer rules |
| list_eventsC | List system events |
| list_alertsC | List system alerts |
| list_capacityC | List system capacity information |
| list_async_jobsC | List asynchronous jobs |
| list_usage_recordsC | List usage records for billing |
| list_zonesC | List availability zones |
| list_templatesC | List templates |
| list_accountsC | List accounts |
| list_usersC | List users |
| list_domainsC | List domains |
| list_hostsD | List hosts |
| list_clustersC | List clusters |
| list_storage_poolsC | List storage pools |
| list_system_vmsC | List system VMs (console proxy, secondary storage) |
| list_service_offeringsC | List service offerings (compute plans) |
| list_ssh_key_pairsC | List SSH key pairs |
| create_ssh_key_pairC | Create a new SSH key pair |
| list_security_groupsC | List security groups |
| create_security_group_ruleC | Create a security group ingress rule |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 45 tools
Most tools map cleanly to distinct resources and actions, but scale_virtual_machine and change_service_offering_virtual_machine overlap heavily, describing the same operation in two ways. This is the main source of confusion.
The naming convention is predominantly verb_noun in snake_case, with many list_* operations. However, there are informal deviations like associate_ip_address, enable_static_nat, and the duplicate pair scale_virtual_machine vs change_service_offering_virtual_machine.
At 45 tools, the surface is very large and includes many list-only operations that could be consolidated or omitted. The breadth feels more like a full API dump than a curated MCP interface.
The surface covers many read operations but has notable gaps: no delete for volumes, snapshots, SSH key pairs, firewall rules, or networks, and no update methods for most resources. This leaves lifecycle workflows incomplete.