> ## Documentation Index
> Fetch the complete documentation index at: https://docs.neoagent.io/llms.txt
> Use this file to discover all available pages before exploring further.

# Scale Computing HyperCore API

> Read and work a client's HyperCore cluster: each VM's vCPU, memory and snapshot schedule, power, snapshots, clones, nodes and cluster health

The tool reaches a client's own HyperCore cluster through its REST API, signed in with the HyperCore user you save for that client, so each call is limited twice: by that user's roles in HyperCore and by the agent's permission groups below.

> [!NOTE]
>
> Automatically enabled when you configure Scale Computing HyperCore permissions in your agent workflow. No manual toggle needed.

## What It Does

* List every VM with its vCPU, memory, disks, NICs, power state and assigned snapshot schedule, and the VMs each schedule protects
* Start, shut down, reboot, reset, pause or live-migrate a VM, and read that it reached the new state
* Take and delete snapshots, create, change and assign snapshot schedules, and start replication
* Clone a VM or a snapshot, add vCPU or memory, add or grow a disk, add a NIC, insert an ISO, create and delete VMs
* Read nodes and their load, drives and their health, cluster conditions, software updates and the state of every task
* Read and change cluster settings, alert targets and users, with the access and approvals you choose

The agent calls the API through the [Vendor API](https://docs.neoagent.io/agents/tools/vendor-api.md) tool, with `scale_hypercore` as the vendor id. [Sandbox](https://docs.neoagent.io/agents/tools/built-in/sandbox.md) scripts can call HyperCore too, under the same permissions. Every change runs as a HyperCore task; the agent reads the task until it ends before it reports or sends the next change.

## Permission Groups

| Group | Covers |
| - | - |
| Virtual Machines | VMs with their vCPU, memory, disks, NICs, state and snapshot schedule, VM stats; create, edit, clone, import, export and delete VMs, and add, change and remove disks and NICs |
| VM Power & Migration | Start, shut down, power off, pause, reboot and reset a VM, and live migration to another node |
| Snapshots | VM snapshots and snapshot schedules, and assigning a schedule to a VM or removing it |
| Replication | VM replication to another cluster and the connections to other clusters, any snapshot or snapshot schedule that turns replication on, and any schedule change that sets replication |
| ISOs & Disk Images | ISO records and uploaded virtual disk images; attach an image to a VM |
| Cluster, Nodes & Tasks | The cluster, nodes, drives, GPUs, conditions, tasks, the API's health, software updates and registration data; shut down the cluster and apply an update |
| Cluster Settings & Alerts | The cluster's name, DNS, time sources and time zone, SMTP relay, alert email and syslog targets, monitoring, SNMP, a node's certificate and the registration with Scale Computing |
| Users & Sign-in | The cluster's users and their roles, and OpenID Connect sign-in |

## Access Profiles

| Profile | What runs on its own | What asks a technician |
| - | - | - |
| **Read Only** | Every read | No write is allowed |
| **Helpdesk** | Every read | Snapshots and snapshot schedules, power and every other VM change. Every other group stays read only |
| **IT Admin** | Every read and every write but the next column | Cluster settings and alert targets (where the cluster's alerts, logs and metrics go), and users and sign-in |
| **Full Automation** | Every read and write | Only the writes that always ask (below) |

## Safety Controls

| Control | Behavior |
| - | - |
| **Always ask** | These always need a technician's approval, whatever the profile: deleting a VM, a disk, a snapshot or a snapshot schedule; exporting a VM to a share or importing one from it; shutting down the cluster; applying a software update; replacing a node's certificate; changing the cluster's registration with Scale Computing; adding, changing or removing anywhere the cluster sends data: an alert email or syslog target, the SMTP relay, monitoring, SNMP, DNS and time servers, and the OpenID Connect provider; creating, changing or deleting a user; setting cloud-init data, which the VM runs when it boots; and a write that sends a password, secret, private key, SNMP community or a share address with a password in it |
| **Hidden credentials** | Passwords, secrets, private keys, the SNMP community and the password inside a share address read `[hidden by Neo]` in every answer. Neo refuses a write that sends the marker; a change that leaves the field out keeps the stored value |
| **One cluster per client** | A call reaches the cluster of the client it names, which on a ticket is the ticket's company |
| **Never sent** | Neo signs every call itself and opens no session, and does not upload ISO or disk image files |
| **Your cluster** | Neo sends the password only over HTTPS to the node address you saved, which must be a public address with a trusted certificate, and never follows a redirect |


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.