Manage an instance
Understand what the selected Mind is running and confirm changes in its state.
An instance is the unit you connect to, operate, and monitor. Open Manage for the selected Mind to inspect the controls available in the current release.
Before changing a Mind
Confirm the instance name, organization, endpoint, and current state. For an operation that can interrupt work, consider the applications and agents using the instance and the data recovery path you need.
A lifecycle request can be accepted before infrastructure is ready. Confirm the resulting state through a fresh management read and an authenticated instance request. If an operation fails, preserve its request identifier and check for a partial result before submitting again.
Free and Pro
Both plans have the same dedicated Firecracker VM–based Akasha requirement. Free caps usage. Pro is intended to expand capacity on the same Mind. Plan differences should not silently turn a full engine capability into a feature wall.
The latest local customer flow delegates to the new instance service, replacing the earlier Render path. The coordinator reports the hosted application still uses the earlier path; the deployed Firecracker lifecycle and effective Free-to-Pro resizing remain unverified. See the instance model and provisioning API.
Controls that need additional verification
In the initial audit, Snapshot and Open logs handlers displayed success text without completing those operations. A UI correction has been reported deployed; actual snapshot creation and log export remain unverified. Do not treat those messages as a backup or an exported log. Use only a verified backend operation and confirm its result. See backups and recovery.