Struct AgentContext 

pub struct AgentContext(/* private fields */);

AI Agent conversation context.

Reference: https://open.longbridge.com/en/docs/ai/chat/conversation

Implementations

impl AgentContext

pub fn new(config: Arc<Config>) -> Self

Create an AgentContext

pub fn log_subscriber(&self) -> Arc<dyn Subscriber + Send + Sync>

Returns the log subscriber

pub async fn workspaces(&self) -> Result<WorkspacesResponse>

List the Workspaces the current account belongs to.

Path: GET /v1/ai/workspaces

pub async fn agents( &self, workspace_id: impl Into<String>, opts: impl Into<Option<GetAgentsOptions>>, ) -> Result<AgentsResponse>

List the Agents in the specified Workspace.

Path: GET /v1/ai/workspaces/{id}/agents

pub async fn public_agents( &self, opts: impl Into<Option<GetAgentsOptions>>, ) -> Result<AgentsResponse>

List all publicly available Agents on the platform — the same catalog shown on the Explore page.

Unlike agents, this endpoint is not scoped to a Workspace: it returns every Agent that is published and publicly shared. The returned Agent::uid is used as the path parameter of conversation.

Path: GET /v1/ai/agents

pub async fn conversation( &self, agent_id: impl Into<String>, query: impl Into<String>, chat_uid: impl Into<Option<String>>, parent_message_id: impl Into<Option<String>>, ) -> Result<ConversationResponse>

Start a conversation with the specified Agent, blocking until the run succeeds, is interrupted, or fails.

parent_message_id is the message_id from a previous response. Pass it when asking a follow-up in an existing conversation to attach the new message after the specified one, keeping the message stream in order. It is only valid together with chat_uid, the parent message must belong to that conversation, and it must not be set for a new conversation.

Path: POST /v1/ai/agents/{id}/conversations

pub async fn continue_conversation( &self, agent_id: impl Into<String>, chat_uid: impl Into<String>, message_id: impl Into<String>, answers: AnswersByToolCall, ) -> Result<ConversationResponse>

Resume an interrupted conversation, blocking until the run succeeds, is interrupted again, or fails.

Path: POST /v1/ai/agents/{id}/conversations/{chat_uid}/messages/{message_id}/ continue

pub async fn conversation_streamed( &self, agent_id: impl Into<String>, query: impl Into<String>, chat_uid: impl Into<Option<String>>, parent_message_id: impl Into<Option<String>>, ) -> Result<impl Stream<Item = Result<ConversationStreamEvent>> + Send + 'static>

Start a conversation with the specified Agent, returning a [Stream] of run-progress events over SSE. The run’s outcome is carried by a ConversationStreamEvent::WorkflowFinished event (succeeded, failed, or stopped) or, if the Agent needs more input from you, a ConversationStreamEvent::HumanInteractionRequired event instead — an interrupted run never emits WorkflowFinished. Neither is necessarily the last item — the server may still emit a few more housekeeping events (e.g. ConversationStreamEvent::ChatTitleUpdated) before actually closing the connection, so keep draining the stream until it ends rather than stopping as soon as you see one.

parent_message_id behaves as in Self::conversation.

Path: POST /v1/ai/agents/{id}/conversations (Accept: text/event-stream)

pub async fn continue_conversation_streamed( &self, agent_id: impl Into<String>, chat_uid: impl Into<String>, message_id: impl Into<String>, answers: AnswersByToolCall, ) -> Result<impl Stream<Item = Result<ConversationStreamEvent>> + Send + 'static>

Resume an interrupted conversation, returning a [Stream] of run-progress events over SSE.

Path: POST /v1/ai/agents/{id}/conversations/{chat_uid}/messages/{message_id}/ continue (Accept: text/event-stream)

Trait Implementations

impl Clone for AgentContext

fn clone(&self) -> AgentContext

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.