Studio

3 entities and 14 agent-callable operations in studio. Each operation is addressed as {mega_tool}/studio/{entity}/{operation} and carries the same permission check over MCP and REST.

environment

OperationMega-toolWhat it does
deprovisionagent_commandTear down the caller's own Studio: stop+remove seat and session containers, disable the public tunnel, soft-delete the row.
modifyagent_commandWhitelist write on the caller's own Studio: default_login_mode and/or login_mode_overrides only.
provisionagent_commandProvision the caller's own tenant Studio (profile forced to 'standard').
publishagent_commandPublish a developer Studio's public app (dev-profile only).
restartagent_commandRestart the caller's own Studio container (stop + start).
resumeagent_commandWarm-resume the caller's own suspended Studio.
statusagent_commandStatus of the caller's own Studio (state, gate, login-mode, published, seat summary).
suspendagent_commandWarm-pause the caller's own Studio (workspace retained).
unpublishagent_commandTake a developer Studio's public app offline (dev-profile only).

guide

OperationMega-toolWhat it does
getagent_commandStudio self-service provisioning playbook.

seat

OperationMega-toolWhat it does
addagent_commandProvision a blank zero-trust seat for a user in the caller's tenant.
listagent_commandList the per-user seats on the caller's own Studio.
removeagent_commandStop + remove a seat's container and soft-delete the seat row.
set_chat_launchagent_commandToggle a seat's opt-in to PROPOSE agent mission launches from chat (the human still approves each proposal).

Parameter schemas and permissions

Parameter schemas and permission codenames are in the full reference — sign in — or call prism_introspect(dimension="capability_search", query="…") with your key.