Inputs

What each operation takes: its path and query parameters and its body, as one object.

Properties

Property	Type	Description
`agents.list`	`object`	-
`agents.list.agents?`	`string`[]	Agent IDs, each listed once. Each includes its conversations. Defaults to every agent you can access.
`agents.list.after?`	`string`	Return results after this cursor. Use the next value from the previous page.
`agents.list.limit?`	`number`	The maximum number to return. Defaults to all.
`agents.forget`	`object`	-
`agents.forget.agents?`	`string`[]	Agent IDs, each listed once. Each includes its conversations. Defaults to every agent you can access.
`interrupts.list`	`object`	-
`interrupts.list.agents?`	`string`[]	Agent IDs, each listed once. Each includes its conversations. Defaults to every agent you can access.
`events.list`	`object`	-
`events.list.agents`	`string`[]	Agent IDs to read, each listed once. Each includes its conversations.
`events.list.types?`	`string`[]	The event types to return, for example wardrik.run.completed. List each type once. Defaults to all types. wardrik.events.dropped is always returned.
`events.list.after?`	`string`	Return events after this cursor. Use the next value from the previous page, an event ID from a stream, or the cursor from agent.read, agent.timeline or agent.runs.
`events.list.before?`	`string`	With order newest, return events before this cursor. Use the next value from the previous page.
`events.list.order?`	`"oldest"` \| `"newest"`	oldest returns the oldest events first, after the cursor; newest returns the newest first, before the cursor or from the end. Defaults to oldest.
`events.list.limit?`	`number`	The maximum number to return. Defaults to 100.
`events.head`	`object`	-
`events.head.agents`	`string`[]	Agent IDs to read, each listed once. Each includes its conversations.
`agent.read`	`object`	-
`agent.read.agent`	`string`	The agent ID, URL-encoded as one path segment.
`agent.read.conversation?`	`string`	The conversation to act on. The agent's own history where left out.
`agent.define`	`object` & `AgentDefinition`	-
`agent.forget`	`object`	-
`agent.forget.agent`	`string`	The agent ID, URL-encoded as one path segment.
`agent.ask`	`object` & `Ask1` & `Ask2`	-
`agent.decide`	`object` & `Decisions`	-
`agent.authorization`	`object`	-
`agent.authorization.agent`	`string`	The agent ID, URL-encoded as one path segment.
`agent.authorization.interrupt`	`string`	The ID of the pending tool call.
`agent.authorization.conversation?`	`string`	The conversation to act on. The agent's own history where left out.
`agent.cancel`	`object`	-
`agent.cancel.agent`	`string`	The agent ID, URL-encoded as one path segment.
`agent.cancel.conversation?`	`string`	The conversation to act on. The agent's own history where left out.
`agent.events`	`object`	-
`agent.events.agent`	`string`	The agent ID, URL-encoded as one path segment.
`agent.events.types?`	`string`[]	The event types to return, for example wardrik.run.completed. List each type once. Defaults to all types. wardrik.events.dropped is always returned.
`agent.events.after?`	`string`	Return events after this cursor. Use the next value from the previous page, an event ID from a stream, or the cursor from agent.read, agent.timeline or agent.runs.
`agent.events.before?`	`string`	With order newest, return events before this cursor. Use the next value from the previous page.
`agent.events.order?`	`"oldest"` \| `"newest"`	oldest returns the oldest events first, after the cursor; newest returns the newest first, before the cursor or from the end. Defaults to oldest.
`agent.events.limit?`	`number`	The maximum number to return. Defaults to 100.
`agent.events.conversation?`	`string`	The conversation to act on. The agent's own history where left out.
`agent.timeline`	`object`	-
`agent.timeline.agent`	`string`	The agent ID, URL-encoded as one path segment.
`agent.timeline.conversation?`	`string`	The conversation to act on. The agent's own history where left out.
`agent.runs`	`object`	-
`agent.runs.agent`	`string`	The agent ID, URL-encoded as one path segment.
`agent.runs.conversation?`	`string`	The conversation to act on. The agent's own history where left out.
`audit.list`	`object`	-
`audit.list.starting_after?`	`number`	Return events after this sequence number. Use the next value from the previous page.
`audit.list.limit?`	`number`	The maximum number to return. Defaults to 100.
`agent.approve`	`object` & `Approve`	-
`agent.deny`	`object` & `Deny`	-
`agents.cancel`	`object`	-
`agents.cancel.agents?`	`string`[]	Agent IDs, each listed once. Each includes its conversations. Defaults to every agent you can access.
`agent.definition`	`object`	-
`agent.definition.agent`	`string`	The agent ID, URL-encoded as one path segment.
`agent.spawn`	`object`	-
`agent.spawn.agent`	`string`	The agent ID, URL-encoded as one path segment.
`agent.stop`	`object`	-
`agent.stop.agent`	`string`	The agent ID, URL-encoded as one path segment.
`agent.stop.conversation?`	`string`	The conversation to act on. The agent's own history where left out.
`agent.clear`	`object`	-
`agent.clear.agent`	`string`	The agent ID, URL-encoded as one path segment.
`agent.clear.conversation?`	`string`	The conversation to act on. The agent's own history where left out.
`agent.calls`	`object`	-
`agent.calls.agent`	`string`	The agent ID, URL-encoded as one path segment.
`agent.calls.conversation?`	`string`	The conversation to act on. The agent's own history where left out.
`agent.runtime`	`object`	-
`agent.runtime.agent`	`string`	The agent ID, URL-encoded as one path segment.
`agent.resolved`	`object`	-
`agent.resolved.agent`	`string`	The agent ID, URL-encoded as one path segment.
`agent.files`	`object`	-
`agent.files.agent`	`string`	The agent ID, URL-encoded as one path segment.
`agent.files.glob?`	`string`	Return only the paths this glob matches: * within a folder, ? one character, across folders. Defaults to .
`agent.file`	`object`	-
`agent.file.agent`	`string`	The agent ID, URL-encoded as one path segment.
`agent.file.path`	`string`	The file's path in the workspace, such as notes/today.md.
`agent.file_put`	`object` & `AgentFilePut`	-
`agent.file_delete`	`object`	-
`agent.file_delete.agent`	`string`	The agent ID, URL-encoded as one path segment.
`agent.file_delete.path`	`string`	The file's path in the workspace, such as notes/today.md.
`agent.kill`	`object`	-
`agent.kill.agent`	`string`	The agent ID, URL-encoded as one path segment.
`agent.kill.call`	`string`	The tool call ID.
`agent.kill.conversation?`	`string`	The conversation to act on. The agent's own history where left out.
`agent.trigger_set`	`object` & `TriggerDefinition`	-
`agent.trigger_remove`	`object`	-
`agent.trigger_remove.agent`	`string`	The agent ID, URL-encoded as one path segment.
`agent.trigger_remove.trigger`	`string`	The trigger name.
`deployment.reload`	`object`	-
`deployment.reload.agents?`	`string`[]	Agent IDs, each listed once. Each includes its conversations. Defaults to every agent you can access.
`deployment.nodes`	`object`	-
`deployment.runtime`	`object`	-
`deployment.runtime.limit?`	`number`	The number of agents to return. Defaults to 20.
`agent.move`	`object` & `Node`	-