Every action below is available as a verb in Cerebral OS — callable from a Cerebral, a Map, or the Runtime API. All executions are governed, audited, and dry-run safe.
Create Agent
chaindesk:create_agent
Create a new AI agent with specified configuration.
Write
Medium risk
Create Datasource
chaindesk:create_datasource
Create a new datasource for training AI agents.
Write
Medium risk
Delete Agent
chaindesk:delete_agent
Permanently delete an agent and all its associated data.
Write
High risk
Delete Datasource
chaindesk:delete_datasource
Permanently delete a datasource and remove it from all agents.
Write
High risk
Get Agent
chaindesk:get_agent
Fetch a single agent by ID with configuration and settings.
Read
Low risk
Get Conversation
chaindesk:get_conversation
Fetch a single conversation by ID with optional message history.
Read
Low risk
Get Datasource
chaindesk:get_datasource
Fetch a single datasource by ID with configuration and status.
Read
Low risk
List Agents
chaindesk:list_agents
List all agents with optional filtering and pagination.
Read
Low risk
List Conversations
chaindesk:list_conversations
List conversations with optional filtering and pagination.
Read
Low risk
List Datasources
chaindesk:list_datasources
List all datasources with optional filtering and pagination.
Read
Low risk
Send Message
chaindesk:send_message
Send a message to an agent and get AI response.
Write
Medium risk
Sync Datasource
chaindesk:sync_datasource
Trigger manual sync/reprocessing of a datasource.
Write
Medium risk
Update Agent
chaindesk:update_agent
Update an existing agent's configuration.
Write
Medium risk
Update Datasource
chaindesk:update_datasource
Update an existing datasource configuration.
Write
Medium risk