@arizeai/phoenix-client 7.4.0 → 7.5.0

package/dist/src/__generated__/api/v1.d.ts

@@ -1660,30 +1660,84 @@ export interface paths {
         put?: never;
         /**
          * Run Server Agent
-         * @description Stream a chat turn from the GraphQL server agent.
-         *
-         *     This is the endpoint the PXI CLI talks to directly (no pre-configured
-         *     agent record): it builds a fresh server agent per request from the
-         *     caller-supplied model and contexts, then streams the reply back as
-         *     Vercel-AI chunks.
-         *
-         *     The request contexts gate capabilities — GraphQL mutations, web access,
-         *     and subagents — and mutations are refused for viewer users. When trace
-         *     recording is enabled (and permitted by system settings), the run is
-         *     traced; locally ingested traces are persisted to the agent's project
-         *     once the stream completes.
-         *
-         *     Returns ``403`` if agents or the server agent are disabled, or if a
-         *     viewer requests mutations.
+         * @deprecated
+         */
+        post: operations["legacyServerAgentChat"];
+        delete?: never;
+        options?: never;
+        head?: never;
+        patch?: never;
+        trace?: never;
+    };
+    "/v1/agent_sessions": {
+        parameters: {
+            query?: never;
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        /**
+         * List Sessions
+         * @description List the viewer's persisted sessions, most recently active first.
+         */
+        get: operations["listAgentSessions"];
+        put?: never;
+        /**
+         * Create Session
+         * @description Create a persisted agent session owned by the requesting user.
+         */
+        post: operations["createAgentSession"];
+        delete?: never;
+        options?: never;
+        head?: never;
+        patch?: never;
+        trace?: never;
+    };
+    "/v1/agent_sessions/{session_id}": {
+        parameters: {
+            query?: never;
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        /**
+         * Get Session
+         * @description Retrieve an owned session's metadata.
+         */
+        get: operations["getAgentSession"];
+        put?: never;
+        post?: never;
+        delete?: never;
+        options?: never;
+        head?: never;
+        /**
+         * Patch Session
+         * @description Update a persisted session's mutable fields.
+         */
+        patch: operations["patchAgentSession"];
+        trace?: never;
+    };
+    "/v1/agent_sessions/{session_id}/messages": {
+        parameters: {
+            query?: never;
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        /**
+         * List Session Messages
+         * @description Page through an owned session's persisted transcript, oldest first.
          */
-        post: operations["run_server_agent_agents_server_sessions__session_id__chat_post"];
+        get: operations["listAgentSessionMessages"];
+        put?: never;
+        post?: never;
         delete?: never;
         options?: never;
         head?: never;
         patch?: never;
         trace?: never;
     };
-    "/agents/{agent_id}/sessions/{session_id}/chat": {
+    "/v1/agent_sessions/{session_id}/compact": {
         parameters: {
             query?: never;
             header?: never;
@@ -1692,15 +1746,35 @@ export interface paths {
         };
         get?: never;
         put?: never;
-        /** Chat */
-        post: operations["chat_agents__agent_id__sessions__session_id__chat_post"];
+        /** Compact Agent Session */
+        post: operations["compactAgentSession"];
+        delete?: never;
+        options?: never;
+        head?: never;
+        patch?: never;
+        trace?: never;
+    };
+    "/v1/agent_sessions/{session_id}/tool_outputs": {
+        parameters: {
+            query?: never;
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        get?: never;
+        put?: never;
+        /**
+         * Submit Agent Session Tool Outputs
+         * @description Persist resolved client tool outputs for the session's open turn.
+         */
+        post: operations["submitAgentSessionToolOutputs"];
         delete?: never;
         options?: never;
         head?: never;
         patch?: never;
         trace?: never;
     };
-    "/agents/{agent_id}/sessions/{session_id}/summary": {
+    "/v1/agent_sessions/{session_id}/chat": {
         parameters: {
             query?: never;
             header?: never;
@@ -1709,8 +1783,8 @@ export interface paths {
         };
         get?: never;
         put?: never;
-        /** Summarize Endpoint */
-        post: operations["summarize_endpoint_agents__agent_id__sessions__session_id__summary_post"];
+        /** Chat */
+        post: operations["agentSessionChat"];
         delete?: never;
         options?: never;
         head?: never;
@@ -1776,6 +1850,92 @@ export interface components {
         AddDatasetLabelToDatasetResponseBody: {
             data: components["schemas"]["DatasetLabel"];
         };
+        AgentModelSelection: components["schemas"]["CustomProviderModelSelection"] | components["schemas"]["BuiltInProviderModelSelection"];
+        /**
+         * AgentSessionConflictError
+         * @description Body of every HTTP 409 returned by the agent session routes.
+         *
+         *     - ``agent_session_busy``: another turn holds the session's turn lock.
+         *     - ``agent_session_model_stale``: the request asserted a model the session
+         *       is no longer set to; refetch the session before retrying.
+         *     - ``agent_session_messages_stale``: the send's ``lastMessageId`` no longer
+         *       matches the persisted transcript — another client appended; refetch the
+         *       transcript and retry.
+         *     - ``agent_session_tool_outputs_conflict``: the submitted ``toolOutputs`` do
+         *       not match the transcript's trailing assistant message (no trailing
+         *       assistant message to continue, an unknown ``toolCallId``, or a tool-name
+         *       mismatch). Unlike ``agent_session_messages_stale`` this is not a
+         *       concurrent-writer race but an inconsistent request; fix the client
+         *       rather than retrying.
+         *     - ``agent_session_already_compact``: there are no complete turns to
+         *       compact — either nothing new has finished since the transcript's latest
+         *       checkpoint, or a concurrent request's checkpoint already covers them.
+         *       Not retryable; the conversation is as compact as it can get.
+         *     - ``agent_session_compaction_conflict``: the conversation changed while it
+         *       was being compacted; retry.
+         */
+        AgentSessionConflictError: {
+            /**
+             * Code
+             * @description Machine-readable reason the request conflicted.
+             * @enum {string}
+             */
+            code: "agent_session_busy" | "agent_session_model_stale" | "agent_session_messages_stale" | "agent_session_tool_outputs_conflict" | "agent_session_already_compact" | "agent_session_compaction_conflict";
+            /**
+             * Message
+             * @description Optional human-readable elaboration on the conflict.
+             */
+            message?: string | null;
+        };
+        /** AgentSessionData */
+        AgentSessionData: {
+            /** Id */
+            id: string;
+            /** Title */
+            title: string;
+            /**
+             * Created At
+             * Format: date-time
+             */
+            created_at: string;
+            /**
+             * Updated At
+             * Format: date-time
+             */
+            updated_at: string;
+            /** Is Ephemeral */
+            is_ephemeral: boolean;
+            model: components["schemas"]["AgentModelSelection"];
+            /**
+             * Is Active
+             * @description Whether a response is currently streaming on this session, i.e. its lock has a live (non-stale) heartbeat.
+             */
+            is_active: boolean;
+            /**
+             * Last Message Id
+             * @description The message ID of the most recently persisted transcript message, or null for an empty transcript.
+             */
+            last_message_id?: string | null;
+        };
+        /** AgentSessionSummary */
+        AgentSessionSummary: {
+            /** Id */
+            id: string;
+            /** Title */
+            title: string;
+            /**
+             * Created At
+             * Format: date-time
+             */
+            created_at: string;
+            /**
+             * Updated At
+             * Format: date-time
+             */
+            updated_at: string;
+            /** Is Ephemeral */
+            is_ephemeral: boolean;
+        };
         /**
          * AgentSpanContext
          * @description Span the user has selected.
@@ -1929,36 +2089,17 @@ export interface components {
             /** Data */
             data: components["schemas"]["CategoricalAnnotationConfig"] | components["schemas"]["ContinuousAnnotationConfig"] | components["schemas"]["FreeformAnnotationConfig"];
         };
-        /**
-         * AssistantMessageMetadata
-         * @description Wire schema for the chat stream's `message_metadata` payload.
-         */
-        AssistantMessageMetadata: {
-            /**
-             * @description discriminator enum property added by openapi-typescript
-             * @enum {string}
-             */
-            type: "assistant";
-            /** Sessionid */
-            sessionId: string;
-            trace?: components["schemas"]["AssistantMessageMetadataTraceIds"] | null;
-            turnTraceContext?: components["schemas"]["TurnTraceContext"] | null;
-            usage?: components["schemas"]["AssistantMessageMetadataUsage"] | null;
-        };
-        /** AssistantMessageMetadataTraceIds */
-        AssistantMessageMetadataTraceIds: {
-            /** Traceid */
-            traceId: string;
-            /** Rootspanid */
-            rootSpanId: string;
-        };
         /** AssistantMessageMetadataUsage */
         AssistantMessageMetadataUsage: {
             tokens: components["schemas"]["AssistantMessageMetadataUsageTokens"];
-            promptDetails?: components["schemas"]["AssistantMessageMetadataUsageTokenDetails"] | null;
+            promptDetails?: components["schemas"]["AssistantMessageMetadataUsageCacheTokenDetails"] | null;
         };
-        /** AssistantMessageMetadataUsageTokenDetails */
-        AssistantMessageMetadataUsageTokenDetails: {
+        /**
+         * AssistantMessageMetadataUsageCacheTokenDetails
+         * @description Prompt-cache token counts, mounted as the usage payload's
+         *     ``prompt_details`` because cached tokens are a breakdown of the prompt.
+         */
+        AssistantMessageMetadataUsageCacheTokenDetails: {
             /** Cacheread */
             cacheRead: number;
             /** Cachewrite */
@@ -1979,8 +2120,7 @@ export interface components {
          *
          *     Credentials and connection details (base URL, Azure endpoint, AWS
          *     region) are resolved from the secret store first and the process
-         *     environment second. ``openai_api_type`` is honoured by the OpenAI and
-         *     Azure OpenAI branches; other providers ignore it.
+         *     environment second.
          */
         BuiltInProviderModelSelection: {
             /**
@@ -1991,12 +2131,6 @@ export interface components {
             provider: components["schemas"]["ModelProvider"];
             /** Modelname */
             modelName: string;
-            /**
-             * Openaiapitype
-             * @default responses
-             * @enum {string}
-             */
-            openaiApiType?: "chat_completions" | "responses";
         };
         /** CategoricalAnnotationConfig */
         CategoricalAnnotationConfig: {
@@ -2135,37 +2269,15 @@ export interface components {
          */
         ChatContext: components["schemas"]["AppContext"] | components["schemas"]["ProjectContext"] | components["schemas"]["TraceContext"] | components["schemas"]["SessionContext"] | components["schemas"]["PromptContext"] | components["schemas"]["PromptVersionContext"] | components["schemas"]["AgentSpanContext"] | components["schemas"]["PlaygroundContext"] | components["schemas"]["CodeEvaluatorContext"] | components["schemas"]["LlmEvaluatorContext"] | components["schemas"]["DatasetContext"] | components["schemas"]["GraphQLContext"] | components["schemas"]["WebAccessContext"] | components["schemas"]["SubagentsContext"];
         /**
-         * ChatRegenerateMessage
-         * @description Regenerate message extended with Phoenix-specific fields.
+         * ChatRequestBody
+         * @description Assistant chat submit request payload.
          */
-        ChatRegenerateMessage: {
-            /**
-             * @description discriminator enum property added by openapi-typescript
-             * @enum {string}
-             */
-            trigger: "regenerate-message";
-            /** Id */
-            id: string;
-            /** Messages */
-            messages: components["schemas"]["PhoenixUIMessage"][];
-            /** Messageid */
-            messageId?: string | null;
-            /**
-             * Ingesttraces
-             * @default false
-             */
-            ingestTraces?: boolean;
-            /**
-             * Exportremotetraces
-             * @default false
-             */
-            exportRemoteTraces?: boolean;
+        ChatRequestBody: {
             /**
-             * Attachuserid
-             * @description When true and the request is authenticated as a PhoenixUser, attaches the user's email as the OpenInference ``user.id`` span attribute on all traced work for this request.
-             * @default false
+             * Headless
+             * @description Whether a headless client (terminal or scripted) is driving the turn, as opposed to the browser assistant. Selects the agent configuration the turn runs on.
              */
-            attachUserId?: boolean;
+            headless: boolean;
             /** Contexts */
             contexts?: components["schemas"]["ChatContext"][];
             /**
@@ -2179,65 +2291,44 @@ export interface components {
              * @description Skills the user explicitly requested via the prompt's slash-command affordance. The server force-loads each available skill by injecting a synthetic load_skill tool call/result at the tail of the message history. Unknown or context-unavailable names are ignored.
              */
             requestedSkills?: string[];
-            /** Model */
-            model: components["schemas"]["CustomProviderModelSelection"] | components["schemas"]["BuiltInProviderModelSelection"];
-            turnTraceContext?: components["schemas"]["TurnTraceContext"] | null;
-        } & {
-            [key: string]: unknown;
-        };
-        /**
-         * ChatRequest
-         * @description Discriminated union of chat request payloads.
-         */
-        ChatRequest: components["schemas"]["ChatSubmitMessage"] | components["schemas"]["ChatRegenerateMessage"];
-        /**
-         * ChatSubmitMessage
-         * @description Submit message extended with Phoenix-specific fields.
-         */
-        ChatSubmitMessage: {
+            /** @description The model the client believes the session is set to. This is a precondition, not an instruction: the turn always runs on the session's persisted selection, and a mismatch is rejected with HTTP 409 and code ``agent_session_model_stale`` rather than silently running on — or switching to — an unexpected model. Change the session's model with ``PATCH .../agent_sessions/{session_id}``. */
+            model: components["schemas"]["AgentModelSelection"];
             /**
-             * @description discriminator enum property added by openapi-typescript
-             * @enum {string}
+             * Trigger
+             * @default submit-message
+             * @constant
              */
-            trigger: "submit-message";
+            trigger?: "submit-message";
             /** Id */
             id: string;
-            /** Messages */
-            messages: components["schemas"]["PhoenixUIMessage"][];
+            /** @description The turn's new user message to append. May be omitted for client-tool continuation, where ``toolOutputs`` resolve the trailing assistant message's pending tool calls instead. */
+            message?: components["schemas"]["PhoenixUIMessage"] | null;
             /**
-             * Ingesttraces
+             * Tooloutputs
+             * @description Client-executed tool results for pending tool calls on the transcript's trailing assistant message, matched by ``toolCallId``. Submitted alone they continue the assistant turn; submitted with ``message`` they resolve dangling tool calls before the new user turn runs.
+             */
+            toolOutputs?: (components["schemas"]["phoenix__db__types__data_stream_protocol__request_types__ToolOutputAvailablePart"] | components["schemas"]["phoenix__db__types__data_stream_protocol__request_types__ToolOutputErrorPart"] | components["schemas"]["phoenix__db__types__data_stream_protocol__request_types__DynamicToolOutputAvailablePart"] | components["schemas"]["phoenix__db__types__data_stream_protocol__request_types__DynamicToolOutputErrorPart"])[];
+            /**
+             * Lastmessageid
+             * @description The id of the last transcript message the client has rendered, used for optimistic concurrency. Omit when the session has no messages; required (and validated against the persisted transcript) once it does. On mismatch the server rejects the send with HTTP 409 and code ``agent_session_messages_stale`` — the client should refetch the session before retrying.
+             */
+            lastMessageId?: string | null;
+            /**
+             * Recordlocaltraces
              * @default false
              */
-            ingestTraces?: boolean;
+            recordLocalTraces?: boolean;
             /**
              * Exportremotetraces
              * @default false
              */
             exportRemoteTraces?: boolean;
             /**
-             * Attachuserid
+             * Instrumentuserid
              * @description When true and the request is authenticated as a PhoenixUser, attaches the user's email as the OpenInference ``user.id`` span attribute on all traced work for this request.
              * @default false
              */
-            attachUserId?: boolean;
-            /** Contexts */
-            contexts?: components["schemas"]["ChatContext"][];
-            /**
-             * Editpermission
-             * @default manual
-             * @enum {string}
-             */
-            editPermission?: "manual" | "bypass";
-            /**
-             * Requestedskills
-             * @description Skills the user explicitly requested via the prompt's slash-command affordance. The server force-loads each available skill by injecting a synthetic load_skill tool call/result at the tail of the message history. Unknown or context-unavailable names are ignored.
-             */
-            requestedSkills?: string[];
-            /** Model */
-            model: components["schemas"]["CustomProviderModelSelection"] | components["schemas"]["BuiltInProviderModelSelection"];
-            turnTraceContext?: components["schemas"]["TurnTraceContext"] | null;
-        } & {
-            [key: string]: unknown;
+            instrumentUserId?: boolean;
         };
         /**
          * CodeEvaluatorContext
@@ -2252,6 +2343,23 @@ export interface components {
             /** Evaluatornodeid */
             evaluatorNodeId?: string | null;
         };
+        /**
+         * CompactAgentSessionRequestBody
+         * @description Request a model-generated checkpoint for a persisted conversation.
+         */
+        CompactAgentSessionRequestBody: {
+            /** @description The model the client believes the session is set to. As on the chat route this is a precondition: the summary is generated with the session's persisted selection, and a mismatch is rejected with HTTP 409 and code ``agent_session_model_stale``. */
+            model: components["schemas"]["AgentModelSelection"];
+        };
+        /**
+         * CompactAgentSessionResponseBody
+         * @description The checkpoint message this request created. A 200 always means a new
+         *     checkpoint was persisted; every other outcome is an HTTP 409 whose body's
+         *     ``code`` says why (see ``AgentSessionConflictError``).
+         */
+        CompactAgentSessionResponseBody: {
+            data: components["schemas"]["PhoenixUIMessage"];
+        };
         /** ContinuousAnnotationConfig */
         ContinuousAnnotationConfig: {
             /** Name */
@@ -2288,6 +2396,29 @@ export interface components {
             /** Upper Bound */
             upper_bound?: number | null;
         };
+        /**
+         * CreateAgentSessionRequestBody
+         * @description Request body for creating a persisted agent session.
+         */
+        CreateAgentSessionRequestBody: {
+            model: components["schemas"]["AgentModelSelection"];
+            /**
+             * Title
+             * @description Optional initial title.
+             * @default
+             */
+            title?: string;
+            /**
+             * Is Ephemeral
+             * @description Whether the session should expire after a period of inactivity.
+             * @default false
+             */
+            is_ephemeral?: boolean;
+        };
+        /** CreateAgentSessionResponseBody */
+        CreateAgentSessionResponseBody: {
+            data: components["schemas"]["CreatedAgentSession"];
+        };
         /** CreateAnnotationConfigData */
         CreateAnnotationConfigData: components["schemas"]["CategoricalAnnotationConfigData"] | components["schemas"]["ContinuousAnnotationConfigData"] | components["schemas"]["FreeformAnnotationConfigData"];
         /** CreateAnnotationConfigResponseBody */
@@ -2588,6 +2719,14 @@ export interface components {
             /** Data */
             data: components["schemas"]["LocalUser"] | components["schemas"]["OAuth2User"] | components["schemas"]["LDAPUser"];
         };
+        /** CreatedAgentSession */
+        CreatedAgentSession: {
+            /**
+             * Id
+             * @description The session's GlobalID — the ``session_id`` the chat route expects.
+             */
+            id: string;
+        };
         /** CreatedApiKey */
         CreatedApiKey: {
             /** Id */
@@ -2950,10 +3089,10 @@ export interface components {
             approval?: components["schemas"]["ToolApprovalRequested"] | components["schemas"]["ToolApprovalResponded"] | null;
         };
         /**
-         * DynamicToolOutputAvailablePart
-         * @description Dynamic tool part in output-available state.
+         * DynamicToolOutputDeniedPart
+         * @description Dynamic tool part in output-denied state (tool was denied, terminal state).
          */
-        DynamicToolOutputAvailablePart: {
+        DynamicToolOutputDeniedPart: {
             /**
              * Type
              * @default dynamic-tool
@@ -2968,14 +3107,12 @@ export interface components {
             title?: string | null;
             /**
              * State
-             * @default output-available
+             * @default output-denied
              * @constant
              */
-            state?: "output-available";
+            state?: "output-denied";
             /** Input */
             input: unknown;
-            /** Output */
-            output: unknown;
             /** Providerexecuted */
             providerExecuted?: boolean | null;
             /** Callprovidermetadata */
@@ -2984,90 +3121,14 @@ export interface components {
                     [key: string]: unknown;
                 };
             } | null;
-            /** Preliminary */
-            preliminary?: boolean | null;
             /** Approval */
             approval?: components["schemas"]["ToolApprovalRequested"] | components["schemas"]["ToolApprovalResponded"] | null;
         };
-        /**
-         * DynamicToolOutputDeniedPart
-         * @description Dynamic tool part in output-denied state (tool was denied, terminal state).
-         */
-        DynamicToolOutputDeniedPart: {
+        /** Experiment */
+        Experiment: {
             /**
-             * Type
-             * @default dynamic-tool
-             * @constant
-             */
-            type?: "dynamic-tool";
-            /** Toolname */
-            toolName: string;
-            /** Toolcallid */
-            toolCallId: string;
-            /** Title */
-            title?: string | null;
-            /**
-             * State
-             * @default output-denied
-             * @constant
-             */
-            state?: "output-denied";
-            /** Input */
-            input: unknown;
-            /** Providerexecuted */
-            providerExecuted?: boolean | null;
-            /** Callprovidermetadata */
-            callProviderMetadata?: {
-                [key: string]: {
-                    [key: string]: unknown;
-                };
-            } | null;
-            /** Approval */
-            approval?: components["schemas"]["ToolApprovalRequested"] | components["schemas"]["ToolApprovalResponded"] | null;
-        };
-        /**
-         * DynamicToolOutputErrorPart
-         * @description Dynamic tool part in output-error state.
-         */
-        DynamicToolOutputErrorPart: {
-            /**
-             * Type
-             * @default dynamic-tool
-             * @constant
-             */
-            type?: "dynamic-tool";
-            /** Toolname */
-            toolName: string;
-            /** Toolcallid */
-            toolCallId: string;
-            /** Title */
-            title?: string | null;
-            /**
-             * State
-             * @default output-error
-             * @constant
-             */
-            state?: "output-error";
-            /** Input */
-            input: unknown;
-            /** Errortext */
-            errorText: string;
-            /** Providerexecuted */
-            providerExecuted?: boolean | null;
-            /** Callprovidermetadata */
-            callProviderMetadata?: {
-                [key: string]: {
-                    [key: string]: unknown;
-                };
-            } | null;
-            /** Approval */
-            approval?: components["schemas"]["ToolApprovalRequested"] | components["schemas"]["ToolApprovalResponded"] | null;
-        };
-        /** Experiment */
-        Experiment: {
-            /**
-             * Id
-             * @description The ID of the experiment
+             * Id
+             * @description The ID of the experiment
              */
             id: string;
             /**
@@ -3290,6 +3351,10 @@ export interface components {
             /** Upper Bound */
             upper_bound?: number | null;
         };
+        /** GetAgentSessionResponseBody */
+        GetAgentSessionResponseBody: {
+            data: components["schemas"]["AgentSessionData"];
+        };
         /** GetAllUserApiKeysResponseBody */
         GetAllUserApiKeysResponseBody: {
             /** Data */
@@ -3546,6 +3611,160 @@ export interface components {
              */
             auth_method: "LDAP";
         };
+        /**
+         * LegacyAssistantMessageMetadata
+         * @description Legacy transcripts predate the ``type`` discriminator, so default it here.
+         */
+        LegacyAssistantMessageMetadata: {
+            /**
+             * Type
+             * @default assistant
+             * @constant
+             */
+            type?: "assistant";
+            /** Sessionid */
+            sessionId: string;
+            turnTraceContext?: components["schemas"]["TurnTraceContext"] | null;
+            usage?: components["schemas"]["AssistantMessageMetadataUsage"] | null;
+            /**
+             * Interrupted
+             * @default false
+             */
+            interrupted?: boolean;
+        } & {
+            [key: string]: unknown;
+        };
+        /**
+         * LegacyAssistantMetadataUIMessage
+         * @description ``UIMessage`` with ``metadata`` narrowed to ``PhoenixAssistantMessageMetadata``.
+         */
+        LegacyAssistantMetadataUIMessage: {
+            /** Id */
+            id: string;
+            /**
+             * Role
+             * @enum {string}
+             */
+            role: "system" | "user" | "assistant";
+            metadata?: components["schemas"]["LegacyAssistantMessageMetadata"] | null;
+            /** Parts */
+            parts: (components["schemas"]["TextUIPart"] | components["schemas"]["ReasoningUIPart"] | components["schemas"]["ToolInputStreamingPart"] | components["schemas"]["ToolInputAvailablePart"] | components["schemas"]["pydantic_ai__ui__vercel_ai__request_types__ToolOutputAvailablePart"] | components["schemas"]["pydantic_ai__ui__vercel_ai__request_types__ToolOutputErrorPart"] | components["schemas"]["ToolApprovalRequestedPart"] | components["schemas"]["ToolApprovalRespondedPart"] | components["schemas"]["ToolOutputDeniedPart"] | components["schemas"]["DynamicToolInputStreamingPart"] | components["schemas"]["DynamicToolInputAvailablePart"] | components["schemas"]["pydantic_ai__ui__vercel_ai__request_types__DynamicToolOutputAvailablePart"] | components["schemas"]["pydantic_ai__ui__vercel_ai__request_types__DynamicToolOutputErrorPart"] | components["schemas"]["DynamicToolApprovalRequestedPart"] | components["schemas"]["DynamicToolApprovalRespondedPart"] | components["schemas"]["DynamicToolOutputDeniedPart"] | components["schemas"]["SourceUrlUIPart"] | components["schemas"]["SourceDocumentUIPart"] | components["schemas"]["FileUIPart"] | components["schemas"]["DataUIPart"] | components["schemas"]["StepStartUIPart"])[];
+        };
+        /**
+         * LegacyChatRegenerateMessage
+         * @description Regenerate message extended with Phoenix-specific fields.
+         */
+        LegacyChatRegenerateMessage: {
+            /**
+             * @description discriminator enum property added by openapi-typescript
+             * @enum {string}
+             */
+            trigger: "regenerate-message";
+            /** Id */
+            id: string;
+            /** Messages */
+            messages: components["schemas"]["LegacyAssistantMetadataUIMessage"][];
+            /** Messageid */
+            messageId?: string | null;
+            /**
+             * Ingesttraces
+             * @default false
+             */
+            ingestTraces?: boolean;
+            /**
+             * Exportremotetraces
+             * @default false
+             */
+            exportRemoteTraces?: boolean;
+            /**
+             * Attachuserid
+             * @description When true and the request is authenticated as a PhoenixUser, attaches the user's email as the OpenInference ``user.id`` span attribute on all traced work for this request.
+             * @default false
+             */
+            attachUserId?: boolean;
+            /** Contexts */
+            contexts?: components["schemas"]["ChatContext"][];
+            /**
+             * Editpermission
+             * @default manual
+             * @enum {string}
+             */
+            editPermission?: "manual" | "bypass";
+            /**
+             * Requestedskills
+             * @description Skills the user explicitly requested via the prompt's slash-command affordance. Ignored by this legacy route.
+             */
+            requestedSkills?: string[];
+            model: components["schemas"]["AgentModelSelection"];
+        } & {
+            [key: string]: unknown;
+        };
+        /**
+         * LegacyChatRequest
+         * @description Discriminated union of legacy chat request payloads.
+         */
+        LegacyChatRequest: components["schemas"]["LegacyChatSubmitMessage"] | components["schemas"]["LegacyChatRegenerateMessage"];
+        /**
+         * LegacyChatSubmitMessage
+         * @description Submit message extended with Phoenix-specific fields.
+         */
+        LegacyChatSubmitMessage: {
+            /**
+             * @description discriminator enum property added by openapi-typescript
+             * @enum {string}
+             */
+            trigger: "submit-message";
+            /** Id */
+            id: string;
+            /** Messages */
+            messages: components["schemas"]["LegacyAssistantMetadataUIMessage"][];
+            /**
+             * Ingesttraces
+             * @default false
+             */
+            ingestTraces?: boolean;
+            /**
+             * Exportremotetraces
+             * @default false
+             */
+            exportRemoteTraces?: boolean;
+            /**
+             * Attachuserid
+             * @description When true and the request is authenticated as a PhoenixUser, attaches the user's email as the OpenInference ``user.id`` span attribute on all traced work for this request.
+             * @default false
+             */
+            attachUserId?: boolean;
+            /** Contexts */
+            contexts?: components["schemas"]["ChatContext"][];
+            /**
+             * Editpermission
+             * @default manual
+             * @enum {string}
+             */
+            editPermission?: "manual" | "bypass";
+            /**
+             * Requestedskills
+             * @description Skills the user explicitly requested via the prompt's slash-command affordance. Ignored by this legacy route.
+             */
+            requestedSkills?: string[];
+            model: components["schemas"]["AgentModelSelection"];
+        } & {
+            [key: string]: unknown;
+        };
+        /** ListAgentSessionMessagesResponseBody */
+        ListAgentSessionMessagesResponseBody: {
+            /** Data */
+            data: components["schemas"]["PhoenixUIMessage"][];
+            /** Next Cursor */
+            next_cursor: string | null;
+        };
+        /** ListAgentSessionsResponseBody */
+        ListAgentSessionsResponseBody: {
+            /** Data */
+            data: components["schemas"]["AgentSessionSummary"][];
+            /** Next Cursor */
+            next_cursor: string | null;
+        };
         /** ListDatasetExamplesData */
         ListDatasetExamplesData: {
             /** Dataset Id */
@@ -3664,6 +3883,15 @@ export interface components {
             /** Password */
             password?: string;
         };
+        /**
+         * MessageMetadata
+         * @description ``UIMessage.metadata`` as a registry of namespaces.
+         */
+        MessageMetadata: {
+            /** Phoenix */
+            phoenix?: (components["schemas"]["PhoenixAssistantMessageMetadata"] | components["schemas"]["PhoenixUserMessageMetadata"]) | null;
+            pydantic_ai?: components["schemas"]["PydanticAIMessageMetadata"] | null;
+        };
         /**
          * ModelProvider
          * @enum {string}
@@ -3933,6 +4161,23 @@ export interface components {
              */
             message?: string | null;
         };
+        /**
+         * PatchAgentSessionRequestBody
+         * @description Fields to update on a persisted session. Omit a field to leave it unchanged.
+         */
+        PatchAgentSessionRequestBody: {
+            /**
+             * Title
+             * @description New title for the session
+             */
+            title?: string;
+            /** @description New model selection for the session */
+            model?: components["schemas"]["AgentModelSelection"];
+        };
+        /** PatchAgentSessionResponseBody */
+        PatchAgentSessionResponseBody: {
+            data: components["schemas"]["AgentSessionData"];
+        };
         /**
          * PatchPromptRequestBody
          * @description Fields to update on a prompt. Omit a field to leave it unchanged.
@@ -3955,9 +4200,29 @@ export interface components {
         PatchPromptResponseBody: {
             data: components["schemas"]["Prompt"];
         };
+        /**
+         * PhoenixAssistantMessageMetadata
+         * @description The ``phoenix`` metadata namespace of an assistant message.
+         */
+        PhoenixAssistantMessageMetadata: {
+            /**
+             * @description discriminator enum property added by openapi-typescript
+             * @enum {string}
+             */
+            type: "assistant";
+            /** Sessionid */
+            sessionId: string;
+            turnTraceContext?: components["schemas"]["TurnTraceContext"] | null;
+            usage?: components["schemas"]["AssistantMessageMetadataUsage"] | null;
+            /**
+             * Interrupted
+             * @default false
+             */
+            interrupted?: boolean;
+        };
         /**
          * PhoenixUIMessage
-         * @description `UIMessage` with `metadata` narrowed to the Phoenix wire shapes.
+         * @description ``UIMessage`` with metadata narrowed to the Phoenix wire shapes.
          */
         PhoenixUIMessage: {
             /** Id */
@@ -3967,10 +4232,30 @@ export interface components {
              * @enum {string}
              */
             role: "system" | "user" | "assistant";
-            /** Metadata */
-            metadata?: (components["schemas"]["AssistantMessageMetadata"] | components["schemas"]["UserMessageMetadata"]) | null;
+            metadata?: components["schemas"]["MessageMetadata"] | null;
             /** Parts */
-            parts: (components["schemas"]["TextUIPart"] | components["schemas"]["ReasoningUIPart"] | components["schemas"]["ToolInputStreamingPart"] | components["schemas"]["ToolInputAvailablePart"] | components["schemas"]["ToolOutputAvailablePart"] | components["schemas"]["ToolOutputErrorPart"] | components["schemas"]["ToolApprovalRequestedPart"] | components["schemas"]["ToolApprovalRespondedPart"] | components["schemas"]["ToolOutputDeniedPart"] | components["schemas"]["DynamicToolInputStreamingPart"] | components["schemas"]["DynamicToolInputAvailablePart"] | components["schemas"]["DynamicToolOutputAvailablePart"] | components["schemas"]["DynamicToolOutputErrorPart"] | components["schemas"]["DynamicToolApprovalRequestedPart"] | components["schemas"]["DynamicToolApprovalRespondedPart"] | components["schemas"]["DynamicToolOutputDeniedPart"] | components["schemas"]["SourceUrlUIPart"] | components["schemas"]["SourceDocumentUIPart"] | components["schemas"]["FileUIPart"] | components["schemas"]["DataUIPart"] | components["schemas"]["StepStartUIPart"])[];
+            parts: (components["schemas"]["TextUIPart"] | components["schemas"]["ReasoningUIPart"] | components["schemas"]["ToolInputStreamingPart"] | components["schemas"]["ToolInputAvailablePart"] | components["schemas"]["phoenix__db__types__data_stream_protocol__request_types__ToolOutputAvailablePart"] | components["schemas"]["phoenix__db__types__data_stream_protocol__request_types__ToolOutputErrorPart"] | components["schemas"]["ToolApprovalRequestedPart"] | components["schemas"]["ToolApprovalRespondedPart"] | components["schemas"]["ToolOutputDeniedPart"] | components["schemas"]["DynamicToolInputStreamingPart"] | components["schemas"]["DynamicToolInputAvailablePart"] | components["schemas"]["phoenix__db__types__data_stream_protocol__request_types__DynamicToolOutputAvailablePart"] | components["schemas"]["phoenix__db__types__data_stream_protocol__request_types__DynamicToolOutputErrorPart"] | components["schemas"]["DynamicToolApprovalRequestedPart"] | components["schemas"]["DynamicToolApprovalRespondedPart"] | components["schemas"]["DynamicToolOutputDeniedPart"] | components["schemas"]["SourceUrlUIPart"] | components["schemas"]["SourceDocumentUIPart"] | components["schemas"]["FileUIPart"] | components["schemas"]["DataUIPart"] | components["schemas"]["StepStartUIPart"])[];
+        };
+        /**
+         * PhoenixUserMessageMetadata
+         * @description The ``phoenix`` metadata namespace the browser attaches to outgoing
+         *     user messages.
+         */
+        PhoenixUserMessageMetadata: {
+            /**
+             * @description discriminator enum property added by openapi-typescript
+             * @enum {string}
+             */
+            type: "user";
+            /** Currentdatetime */
+            currentDateTime: string;
+            /** Timezone */
+            timeZone: string;
+            /**
+             * Iscompactionmessage
+             * @default false
+             */
+            isCompactionMessage?: boolean;
         };
         /**
          * PlaygroundBuiltinModelContext
@@ -4915,6 +5200,16 @@ export interface components {
                 [key: string]: unknown;
             };
         };
+        /**
+         * PydanticAIMessageMetadata
+         * @description Local pin of pydantic-ai's message-level ``pydantic_ai`` metadata
+         *     namespace (its private ``_PydanticAIMessageMetadata``), merged into the
+         *     assistant metadata by the stream's metadata chunk.
+         */
+        PydanticAIMessageMetadata: {
+            /** Timestamp */
+            timestamp?: string | null;
+        };
         /**
          * ReasoningUIPart
          * @description A reasoning part of a message.
@@ -5544,6 +5839,29 @@ export interface components {
             /** Enabled */
             enabled: boolean;
         };
+        /**
+         * SubmitAgentSessionToolOutputsRequestBody
+         * @description Persist resolved client tool outputs without continuing the turn.
+         */
+        SubmitAgentSessionToolOutputsRequestBody: {
+            /**
+             * Tooloutputs
+             * @description Client tool results for pending calls on the trailing assistant message, matched by ``toolCallId``. Resending a persisted output verbatim is a no-op; an output that differs from the persisted result or matches no call is rejected with HTTP 409 and code ``agent_session_tool_outputs_conflict``.
+             */
+            toolOutputs: (components["schemas"]["phoenix__db__types__data_stream_protocol__request_types__ToolOutputAvailablePart"] | components["schemas"]["phoenix__db__types__data_stream_protocol__request_types__ToolOutputErrorPart"] | components["schemas"]["phoenix__db__types__data_stream_protocol__request_types__DynamicToolOutputAvailablePart"] | components["schemas"]["phoenix__db__types__data_stream_protocol__request_types__DynamicToolOutputErrorPart"])[];
+            /**
+             * Lastmessageid
+             * @description The trailing assistant message's id. On mismatch the submission is rejected with HTTP 409 and code ``agent_session_messages_stale``.
+             */
+            lastMessageId: string;
+        };
+        /**
+         * SubmitAgentSessionToolOutputsResponseBody
+         * @description The trailing assistant message with the submitted outputs applied.
+         */
+        SubmitAgentSessionToolOutputsResponseBody: {
+            data: components["schemas"]["PhoenixUIMessage"];
+        };
         /** TextContentPart */
         TextContentPart: {
             /**
@@ -5741,10 +6059,10 @@ export interface components {
             approval?: components["schemas"]["ToolApprovalRequested"] | components["schemas"]["ToolApprovalResponded"] | null;
         };
         /**
-         * ToolOutputAvailablePart
-         * @description Tool part in output-available state.
+         * ToolOutputDeniedPart
+         * @description Tool part in output-denied state (tool was denied, terminal state).
          */
-        ToolOutputAvailablePart: {
+        ToolOutputDeniedPart: {
             /** Type */
             type: string;
             /** Toolcallid */
@@ -5753,80 +6071,12 @@ export interface components {
             title?: string | null;
             /**
              * State
-             * @default output-available
+             * @default output-denied
              * @constant
              */
-            state?: "output-available";
+            state?: "output-denied";
             /** Input */
             input?: unknown | null;
-            /** Output */
-            output?: unknown | null;
-            /** Providerexecuted */
-            providerExecuted?: boolean | null;
-            /** Callprovidermetadata */
-            callProviderMetadata?: {
-                [key: string]: {
-                    [key: string]: unknown;
-                };
-            } | null;
-            /** Preliminary */
-            preliminary?: boolean | null;
-            /** Approval */
-            approval?: components["schemas"]["ToolApprovalRequested"] | components["schemas"]["ToolApprovalResponded"] | null;
-        };
-        /**
-         * ToolOutputDeniedPart
-         * @description Tool part in output-denied state (tool was denied, terminal state).
-         */
-        ToolOutputDeniedPart: {
-            /** Type */
-            type: string;
-            /** Toolcallid */
-            toolCallId: string;
-            /** Title */
-            title?: string | null;
-            /**
-             * State
-             * @default output-denied
-             * @constant
-             */
-            state?: "output-denied";
-            /** Input */
-            input?: unknown | null;
-            /** Providerexecuted */
-            providerExecuted?: boolean | null;
-            /** Callprovidermetadata */
-            callProviderMetadata?: {
-                [key: string]: {
-                    [key: string]: unknown;
-                };
-            } | null;
-            /** Approval */
-            approval?: components["schemas"]["ToolApprovalRequested"] | components["schemas"]["ToolApprovalResponded"] | null;
-        };
-        /**
-         * ToolOutputErrorPart
-         * @description Tool part in output-error state.
-         */
-        ToolOutputErrorPart: {
-            /** Type */
-            type: string;
-            /** Toolcallid */
-            toolCallId: string;
-            /** Title */
-            title?: string | null;
-            /**
-             * State
-             * @default output-error
-             * @constant
-             */
-            state?: "output-error";
-            /** Input */
-            input?: unknown | null;
-            /** Rawinput */
-            rawInput?: unknown | null;
-            /** Errortext */
-            errorText: string;
             /** Providerexecuted */
             providerExecuted?: boolean | null;
             /** Callprovidermetadata */
@@ -6041,7 +6291,10 @@ export interface components {
              */
             end_time: string;
         };
-        /** TurnTraceContext */
+        /**
+         * TurnTraceContext
+         * @description The trace identity a turn's spans are parented to.
+         */
         TurnTraceContext: {
             /** Traceid */
             traceId: string;
@@ -6053,23 +6306,6 @@ export interface components {
              */
             startedAt: string;
         };
-        /**
-         * UIMessage
-         * @description A message as displayed in the UI by Vercel AI Elements.
-         */
-        UIMessage: {
-            /** Id */
-            id: string;
-            /**
-             * Role
-             * @enum {string}
-             */
-            role: "system" | "user" | "assistant";
-            /** Metadata */
-            metadata?: unknown | null;
-            /** Parts */
-            parts: (components["schemas"]["TextUIPart"] | components["schemas"]["ReasoningUIPart"] | components["schemas"]["ToolInputStreamingPart"] | components["schemas"]["ToolInputAvailablePart"] | components["schemas"]["ToolOutputAvailablePart"] | components["schemas"]["ToolOutputErrorPart"] | components["schemas"]["ToolApprovalRequestedPart"] | components["schemas"]["ToolApprovalRespondedPart"] | components["schemas"]["ToolOutputDeniedPart"] | components["schemas"]["DynamicToolInputStreamingPart"] | components["schemas"]["DynamicToolInputAvailablePart"] | components["schemas"]["DynamicToolOutputAvailablePart"] | components["schemas"]["DynamicToolOutputErrorPart"] | components["schemas"]["DynamicToolApprovalRequestedPart"] | components["schemas"]["DynamicToolApprovalRespondedPart"] | components["schemas"]["DynamicToolOutputDeniedPart"] | components["schemas"]["SourceUrlUIPart"] | components["schemas"]["SourceDocumentUIPart"] | components["schemas"]["FileUIPart"] | components["schemas"]["DataUIPart"] | components["schemas"]["StepStartUIPart"])[];
-        };
         /** UpdateAnnotationConfigResponseBody */
         UpdateAnnotationConfigResponseBody: {
             /** Data */
@@ -6289,21 +6525,6 @@ export interface components {
             expires_at?: string | null;
             user: components["schemas"]["ApiKeyUser"];
         };
-        /**
-         * UserMessageMetadata
-         * @description Wire schema for metadata the browser attaches to outgoing user messages.
-         */
-        UserMessageMetadata: {
-            /**
-             * @description discriminator enum property added by openapi-typescript
-             * @enum {string}
-             */
-            type: "user";
-            /** Currentdatetime */
-            currentDateTime: string;
-            /** Timezone */
-            timeZone: string;
-        };
         /** ValidationError */
         ValidationError: {
             /** Location */
@@ -6331,86 +6552,470 @@ export interface components {
             enabled: boolean;
         };
         /**
-         * _SummarizeRequest
-         * @description Body for POST /agents/{agent_id}/sessions/{session_id}/summary.
-         *
-         *     Carries the Vercel-style messages array; the backend owns the prompt and
-         *     the structured-output tool schema.
+         * DynamicToolOutputAvailablePart
+         * @description Dynamic tool part in output-available state.
          */
-        _SummarizeRequest: {
-            /**
-             * Ingesttraces
-             * @default false
-             */
-            ingestTraces?: boolean;
+        phoenix__db__types__data_stream_protocol__request_types__DynamicToolOutputAvailablePart: {
             /**
-             * Exportremotetraces
-             * @default false
+             * Type
+             * @default dynamic-tool
+             * @constant
              */
-            exportRemoteTraces?: boolean;
+            type?: "dynamic-tool";
+            /** Toolname */
+            toolName: string;
+            /** Toolcallid */
+            toolCallId: string;
+            /** Title */
+            title?: string | null;
             /**
-             * Attachuserid
-             * @description When true and the request is authenticated as a PhoenixUser, attaches the user's email as the OpenInference ``user.id`` span attribute on all traced work for this request.
-             * @default false
+             * State
+             * @default output-available
+             * @constant
              */
-            attachUserId?: boolean;
-            /** Messages */
-            messages: components["schemas"]["UIMessage"][];
-            /** Model */
-            model: components["schemas"]["CustomProviderModelSelection"] | components["schemas"]["BuiltInProviderModelSelection"];
-        };
-        /** _SummarizeResponse */
-        _SummarizeResponse: {
-            /** Summary */
-            summary: string;
+            state?: "output-available";
+            /** Input */
+            input: unknown;
+            /** Output */
+            output: unknown;
+            /** Providerexecuted */
+            providerExecuted?: boolean | null;
+            /** Callprovidermetadata */
+            callProviderMetadata?: {
+                [key: string]: {
+                    [key: string]: unknown;
+                };
+            } | null;
+            /** Resultprovidermetadata */
+            resultProviderMetadata?: {
+                [key: string]: {
+                    [key: string]: unknown;
+                };
+            } | null;
+            /** Preliminary */
+            preliminary?: boolean | null;
+            /** Approval */
+            approval?: components["schemas"]["ToolApprovalRequested"] | components["schemas"]["ToolApprovalResponded"] | null;
         };
         /**
-         * ToolCallCallbackProviderMetadata
-         * @description Shape of the ``phoenix`` namespace the browser returns in
-         *     ``callProviderMetadata`` on resolved tool parts: the server-stamped fields
-         *     plus browser-recorded execution timings.
+         * DynamicToolOutputErrorPart
+         * @description Dynamic tool part in output-error state.
          */
-        ToolCallCallbackProviderMetadata: {
-            /**
-             * Tool Execution Environment
-             * @enum {string}
-             */
-            tool_execution_environment: "client" | "server";
-            /**
-             * Tool Input Emitted At
-             * @default null
-             */
-            tool_input_emitted_at?: string | null;
+        phoenix__db__types__data_stream_protocol__request_types__DynamicToolOutputErrorPart: {
             /**
-             * Client Started At
-             * @default null
+             * Type
+             * @default dynamic-tool
+             * @constant
              */
-            client_started_at?: string | null;
+            type?: "dynamic-tool";
+            /** Toolname */
+            toolName: string;
+            /** Toolcallid */
+            toolCallId: string;
+            /** Title */
+            title?: string | null;
             /**
-             * Client Ended At
-             * @default null
+             * State
+             * @default output-error
+             * @constant
              */
-            client_ended_at?: string | null;
+            state?: "output-error";
+            /** Input */
+            input: unknown;
+            /** Errortext */
+            errorText: string;
+            /** Providerexecuted */
+            providerExecuted?: boolean | null;
+            /** Callprovidermetadata */
+            callProviderMetadata?: {
+                [key: string]: {
+                    [key: string]: unknown;
+                };
+            } | null;
+            /** Resultprovidermetadata */
+            resultProviderMetadata?: {
+                [key: string]: {
+                    [key: string]: unknown;
+                };
+            } | null;
+            /** Approval */
+            approval?: components["schemas"]["ToolApprovalRequested"] | components["schemas"]["ToolApprovalResponded"] | null;
         };
         /**
-         * ToolCallProviderMetadata
-         * @description Payload Phoenix stamps under the ``phoenix`` namespace of Vercel AI
-         *     ``providerMetadata`` on tool-call chunks (``tool-input-start`` and
-         *     ``tool-input-available``).
+         * ToolOutputAvailablePart
+         * @description Tool part in output-available state.
          */
-        ToolCallProviderMetadata: {
+        phoenix__db__types__data_stream_protocol__request_types__ToolOutputAvailablePart: {
+            /** Type */
+            type: string;
+            /** Toolcallid */
+            toolCallId: string;
+            /** Title */
+            title?: string | null;
             /**
-             * Tool Execution Environment
-             * @enum {string}
+             * State
+             * @default output-available
+             * @constant
              */
-            tool_execution_environment: "client" | "server";
+            state?: "output-available";
+            /** Input */
+            input?: unknown | null;
+            /** Output */
+            output?: unknown | null;
+            /** Providerexecuted */
+            providerExecuted?: boolean | null;
+            /** Callprovidermetadata */
+            callProviderMetadata?: {
+                [key: string]: {
+                    [key: string]: unknown;
+                };
+            } | null;
+            /** Resultprovidermetadata */
+            resultProviderMetadata?: {
+                [key: string]: {
+                    [key: string]: unknown;
+                };
+            } | null;
+            /** Preliminary */
+            preliminary?: boolean | null;
+            /** Approval */
+            approval?: components["schemas"]["ToolApprovalRequested"] | components["schemas"]["ToolApprovalResponded"] | null;
+        };
+        /**
+         * ToolOutputErrorPart
+         * @description Tool part in output-error state.
+         */
+        phoenix__db__types__data_stream_protocol__request_types__ToolOutputErrorPart: {
+            /** Type */
+            type: string;
+            /** Toolcallid */
+            toolCallId: string;
+            /** Title */
+            title?: string | null;
             /**
-             * Tool Input Emitted At
-             * @default null
+             * State
+             * @default output-error
+             * @constant
              */
-            tool_input_emitted_at?: string | null;
+            state?: "output-error";
+            /** Input */
+            input?: unknown | null;
+            /** Rawinput */
+            rawInput?: unknown | null;
+            /** Errortext */
+            errorText: string;
+            /** Providerexecuted */
+            providerExecuted?: boolean | null;
+            /** Callprovidermetadata */
+            callProviderMetadata?: {
+                [key: string]: {
+                    [key: string]: unknown;
+                };
+            } | null;
+            /** Resultprovidermetadata */
+            resultProviderMetadata?: {
+                [key: string]: {
+                    [key: string]: unknown;
+                };
+            } | null;
+            /** Approval */
+            approval?: components["schemas"]["ToolApprovalRequested"] | components["schemas"]["ToolApprovalResponded"] | null;
         };
-    };
+        /**
+         * DynamicToolOutputAvailablePart
+         * @description Dynamic tool part in output-available state.
+         */
+        pydantic_ai__ui__vercel_ai__request_types__DynamicToolOutputAvailablePart: {
+            /**
+             * Type
+             * @default dynamic-tool
+             * @constant
+             */
+            type?: "dynamic-tool";
+            /** Toolname */
+            toolName: string;
+            /** Toolcallid */
+            toolCallId: string;
+            /** Title */
+            title?: string | null;
+            /**
+             * State
+             * @default output-available
+             * @constant
+             */
+            state?: "output-available";
+            /** Input */
+            input: unknown;
+            /** Output */
+            output: unknown;
+            /** Providerexecuted */
+            providerExecuted?: boolean | null;
+            /** Callprovidermetadata */
+            callProviderMetadata?: {
+                [key: string]: {
+                    [key: string]: unknown;
+                };
+            } | null;
+            /** Preliminary */
+            preliminary?: boolean | null;
+            /** Approval */
+            approval?: components["schemas"]["ToolApprovalRequested"] | components["schemas"]["ToolApprovalResponded"] | null;
+        };
+        /**
+         * DynamicToolOutputErrorPart
+         * @description Dynamic tool part in output-error state.
+         */
+        pydantic_ai__ui__vercel_ai__request_types__DynamicToolOutputErrorPart: {
+            /**
+             * Type
+             * @default dynamic-tool
+             * @constant
+             */
+            type?: "dynamic-tool";
+            /** Toolname */
+            toolName: string;
+            /** Toolcallid */
+            toolCallId: string;
+            /** Title */
+            title?: string | null;
+            /**
+             * State
+             * @default output-error
+             * @constant
+             */
+            state?: "output-error";
+            /** Input */
+            input: unknown;
+            /** Errortext */
+            errorText: string;
+            /** Providerexecuted */
+            providerExecuted?: boolean | null;
+            /** Callprovidermetadata */
+            callProviderMetadata?: {
+                [key: string]: {
+                    [key: string]: unknown;
+                };
+            } | null;
+            /** Approval */
+            approval?: components["schemas"]["ToolApprovalRequested"] | components["schemas"]["ToolApprovalResponded"] | null;
+        };
+        /**
+         * ToolOutputAvailablePart
+         * @description Tool part in output-available state.
+         */
+        pydantic_ai__ui__vercel_ai__request_types__ToolOutputAvailablePart: {
+            /** Type */
+            type: string;
+            /** Toolcallid */
+            toolCallId: string;
+            /** Title */
+            title?: string | null;
+            /**
+             * State
+             * @default output-available
+             * @constant
+             */
+            state?: "output-available";
+            /** Input */
+            input?: unknown | null;
+            /** Output */
+            output?: unknown | null;
+            /** Providerexecuted */
+            providerExecuted?: boolean | null;
+            /** Callprovidermetadata */
+            callProviderMetadata?: {
+                [key: string]: {
+                    [key: string]: unknown;
+                };
+            } | null;
+            /** Preliminary */
+            preliminary?: boolean | null;
+            /** Approval */
+            approval?: components["schemas"]["ToolApprovalRequested"] | components["schemas"]["ToolApprovalResponded"] | null;
+        };
+        /**
+         * ToolOutputErrorPart
+         * @description Tool part in output-error state.
+         */
+        pydantic_ai__ui__vercel_ai__request_types__ToolOutputErrorPart: {
+            /** Type */
+            type: string;
+            /** Toolcallid */
+            toolCallId: string;
+            /** Title */
+            title?: string | null;
+            /**
+             * State
+             * @default output-error
+             * @constant
+             */
+            state?: "output-error";
+            /** Input */
+            input?: unknown | null;
+            /** Rawinput */
+            rawInput?: unknown | null;
+            /** Errortext */
+            errorText: string;
+            /** Providerexecuted */
+            providerExecuted?: boolean | null;
+            /** Callprovidermetadata */
+            callProviderMetadata?: {
+                [key: string]: {
+                    [key: string]: unknown;
+                };
+            } | null;
+            /** Approval */
+            approval?: components["schemas"]["ToolApprovalRequested"] | components["schemas"]["ToolApprovalResponded"] | null;
+        };
+        /**
+         * AgentErrorChunk
+         * @description Durable data part paired with a standard protocol ``error`` chunk.
+         *
+         *     Standard ``error`` chunks are transient: the reducer routes them to error
+         *     callbacks without adding a message part, so they vanish from persisted
+         *     transcripts. Pairing each one with this ``data-error`` part records the
+         *     error durably in the message for persistence, reload, and subagents.
+         */
+        AgentErrorChunk: {
+            /**
+             * Type
+             * @default data-error
+             * @constant
+             */
+            type?: "data-error";
+            /**
+             * Id
+             * @default null
+             */
+            id?: string | null;
+            data: components["schemas"]["AgentErrorData"];
+            /**
+             * Transient
+             * @default null
+             */
+            transient?: boolean | null;
+        };
+        /**
+         * AgentErrorData
+         * @description Payload of the durable ``data-error`` part persisted for protocol errors.
+         */
+        AgentErrorData: {
+            /** Errortext */
+            errorText: string;
+        };
+        /**
+         * PhoenixToolCallCallbackProviderMetadata
+         * @description Shape of the ``phoenix`` namespace the browser returns in
+         *     ``callProviderMetadata`` on resolved tool parts: the server-stamped fields
+         *     plus browser-recorded execution timings.
+         */
+        PhoenixToolCallCallbackProviderMetadata: {
+            /**
+             * Toolexecutionenvironment
+             * @enum {string}
+             */
+            toolExecutionEnvironment: "client" | "server";
+            /**
+             * Toolinputemittedat
+             * @default null
+             */
+            toolInputEmittedAt?: string | null;
+            /**
+             * Clientstartedat
+             * @default null
+             */
+            clientStartedAt?: string | null;
+            /**
+             * Clientendedat
+             * @default null
+             */
+            clientEndedAt?: string | null;
+            /**
+             * Outcome
+             * @default null
+             */
+            outcome?: "interrupted" | null;
+        };
+        /**
+         * PhoenixToolCallProviderMetadata
+         * @description Payload Phoenix stamps under the ``phoenix`` namespace of Vercel AI
+         *     ``providerMetadata`` on tool-call chunks (``tool-input-start`` and
+         *     ``tool-input-available``).
+         */
+        PhoenixToolCallProviderMetadata: {
+            /**
+             * Toolexecutionenvironment
+             * @enum {string}
+             */
+            toolExecutionEnvironment: "client" | "server";
+            /**
+             * Toolinputemittedat
+             * @default null
+             */
+            toolInputEmittedAt?: string | null;
+        };
+        /**
+         * SessionSummaryChunk
+         * @description Transient ``data-session-summary`` stream chunk: the LLM-generated
+         *     session title, emitted on any turn that starts with the session still
+         *     untitled. Being transient, it reaches the client's ``onData`` callback
+         *     but is never appended to the message parts.
+         *
+         *     See the Vercel AI SDK data stream protocol:
+         *         - Data parts: https://ai-sdk.dev/docs/ai-sdk-ui/stream-protocol#data-parts
+         *         - Transient parts: https://ai-sdk.dev/docs/ai-sdk-ui/streaming-data#transient-data-parts-ephemeral
+         */
+        SessionSummaryChunk: {
+            /**
+             * Type
+             * @default data-session-summary
+             * @constant
+             */
+            type?: "data-session-summary";
+            /**
+             * Id
+             * @default null
+             */
+            id?: string | null;
+            /** Data */
+            data: string;
+            /**
+             * Transient
+             * @default true
+             * @constant
+             */
+            transient?: true;
+        };
+        /**
+         * TranscriptPersistedChunk
+         * @description Confirms that a streamed assistant message is durable.
+         */
+        TranscriptPersistedChunk: {
+            /**
+             * Type
+             * @default data-transcript-persisted
+             * @constant
+             */
+            type?: "data-transcript-persisted";
+            /**
+             * Id
+             * @default null
+             */
+            id?: string | null;
+            data: components["schemas"]["TranscriptPersistedData"];
+            /**
+             * Transient
+             * @default true
+             * @constant
+             */
+            transient?: true;
+        };
+        /** TranscriptPersistedData */
+        TranscriptPersistedData: {
+            /** Messageid */
+            messageId: string;
+        };
+    };
     responses: never;
     parameters: never;
     requestBodies: never;
@@ -11758,7 +12363,7 @@ export interface operations {
             };
         };
     };
-    run_server_agent_agents_server_sessions__session_id__chat_post: {
+    legacyServerAgentChat: {
         parameters: {
             query?: never;
             header?: never;
@@ -11769,7 +12374,7 @@ export interface operations {
         };
         requestBody: {
             content: {
-                "application/json": components["schemas"]["ChatRequest"];
+                "application/json": components["schemas"]["LegacyChatRequest"];
             };
         };
         responses: {
@@ -11782,30 +12387,421 @@ export interface operations {
                     "application/json": unknown;
                 };
             };
-            /** @description Validation Error */
-            422: {
+            /** @description Bad Request */
+            400: {
                 headers: {
                     [name: string]: unknown;
                 };
                 content: {
-                    "application/json": components["schemas"]["HTTPValidationError"];
+                    "text/plain": string;
                 };
             };
-        };
-    };
-    chat_agents__agent_id__sessions__session_id__chat_post: {
-        parameters: {
-            query?: never;
-            header?: never;
-            path: {
-                agent_id: string;
-                session_id: string;
+            /** @description Unauthorized */
+            401: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "text/plain": string;
+                };
             };
-            cookie?: never;
-        };
+            /** @description Forbidden */
+            403: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "text/plain": string;
+                };
+            };
+            /** @description Not Found */
+            404: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "text/plain": string;
+                };
+            };
+            /** @description Validation Error */
+            422: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": components["schemas"]["HTTPValidationError"];
+                };
+            };
+            /** @description Insufficient Storage */
+            507: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "text/plain": string;
+                };
+            };
+        };
+    };
+    listAgentSessions: {
+        parameters: {
+            query?: {
+                /** @description Opaque pagination cursor. */
+                cursor?: string | null;
+                limit?: number;
+            };
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        requestBody?: never;
+        responses: {
+            /** @description Successful Response */
+            200: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": components["schemas"]["ListAgentSessionsResponseBody"];
+                };
+            };
+            /** @description Unauthorized */
+            401: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "text/plain": string;
+                };
+            };
+            /** @description Forbidden */
+            403: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "text/plain": string;
+                };
+            };
+            /** @description Not Found */
+            404: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "text/plain": string;
+                };
+            };
+            /** @description Unprocessable Entity */
+            422: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "text/plain": string;
+                };
+            };
+        };
+    };
+    createAgentSession: {
+        parameters: {
+            query?: never;
+            header?: never;
+            path?: never;
+            cookie?: never;
+        };
+        requestBody: {
+            content: {
+                "application/json": components["schemas"]["CreateAgentSessionRequestBody"];
+            };
+        };
+        responses: {
+            /** @description Successful Response */
+            201: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": components["schemas"]["CreateAgentSessionResponseBody"];
+                };
+            };
+            /** @description Bad Request */
+            400: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "text/plain": string;
+                };
+            };
+            /** @description Unauthorized */
+            401: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "text/plain": string;
+                };
+            };
+            /** @description Forbidden */
+            403: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "text/plain": string;
+                };
+            };
+            /** @description Not Found */
+            404: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "text/plain": string;
+                };
+            };
+            /** @description Unprocessable Entity */
+            422: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "text/plain": string;
+                };
+            };
+            /** @description Insufficient Storage */
+            507: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "text/plain": string;
+                };
+            };
+        };
+    };
+    getAgentSession: {
+        parameters: {
+            query?: never;
+            header?: never;
+            path: {
+                session_id: string;
+            };
+            cookie?: never;
+        };
+        requestBody?: never;
+        responses: {
+            /** @description Successful Response */
+            200: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": components["schemas"]["GetAgentSessionResponseBody"];
+                };
+            };
+            /** @description Unauthorized */
+            401: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "text/plain": string;
+                };
+            };
+            /** @description Forbidden */
+            403: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "text/plain": string;
+                };
+            };
+            /** @description Not Found */
+            404: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "text/plain": string;
+                };
+            };
+            /** @description Validation Error */
+            422: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": components["schemas"]["HTTPValidationError"];
+                };
+            };
+        };
+    };
+    patchAgentSession: {
+        parameters: {
+            query?: never;
+            header?: never;
+            path: {
+                session_id: string;
+            };
+            cookie?: never;
+        };
+        requestBody: {
+            content: {
+                "application/json": components["schemas"]["PatchAgentSessionRequestBody"];
+            };
+        };
+        responses: {
+            /** @description Successful Response */
+            200: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": components["schemas"]["PatchAgentSessionResponseBody"];
+                };
+            };
+            /** @description Bad Request */
+            400: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "text/plain": string;
+                };
+            };
+            /** @description Unauthorized */
+            401: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "text/plain": string;
+                };
+            };
+            /** @description Forbidden */
+            403: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "text/plain": string;
+                };
+            };
+            /** @description Not Found */
+            404: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "text/plain": string;
+                };
+            };
+            /** @description The request conflicts with the session's current state; the body's ``code`` field says how. */
+            409: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": components["schemas"]["AgentSessionConflictError"];
+                };
+            };
+            /** @description Unprocessable Entity */
+            422: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "text/plain": string;
+                };
+            };
+            /** @description Insufficient Storage */
+            507: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "text/plain": string;
+                };
+            };
+        };
+    };
+    listAgentSessionMessages: {
+        parameters: {
+            query?: {
+                /** @description Opaque pagination cursor. */
+                cursor?: string | null;
+                limit?: number;
+            };
+            header?: never;
+            path: {
+                session_id: string;
+            };
+            cookie?: never;
+        };
+        requestBody?: never;
+        responses: {
+            /** @description Successful Response */
+            200: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": components["schemas"]["ListAgentSessionMessagesResponseBody"];
+                };
+            };
+            /** @description Unauthorized */
+            401: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "text/plain": string;
+                };
+            };
+            /** @description Forbidden */
+            403: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "text/plain": string;
+                };
+            };
+            /** @description Not Found */
+            404: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "text/plain": string;
+                };
+            };
+            /** @description Unprocessable Entity */
+            422: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "text/plain": string;
+                };
+            };
+        };
+    };
+    compactAgentSession: {
+        parameters: {
+            query?: never;
+            header?: never;
+            path: {
+                session_id: string;
+            };
+            cookie?: never;
+        };
         requestBody: {
             content: {
-                "application/json": components["schemas"]["ChatRequest"];
+                "application/json": components["schemas"]["CompactAgentSessionRequestBody"];
             };
         };
         responses: {
@@ -11815,7 +12811,150 @@ export interface operations {
                     [name: string]: unknown;
                 };
                 content: {
-                    "application/json": unknown;
+                    "application/json": components["schemas"]["CompactAgentSessionResponseBody"];
+                };
+            };
+            /** @description Bad Request */
+            400: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "text/plain": string;
+                };
+            };
+            /** @description Unauthorized */
+            401: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "text/plain": string;
+                };
+            };
+            /** @description Forbidden */
+            403: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "text/plain": string;
+                };
+            };
+            /** @description Not Found */
+            404: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "text/plain": string;
+                };
+            };
+            /** @description The request conflicts with the session's current state; the body's ``code`` field says how. */
+            409: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": components["schemas"]["AgentSessionConflictError"];
+                };
+            };
+            /** @description Validation Error */
+            422: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": components["schemas"]["HTTPValidationError"];
+                };
+            };
+            /** @description Bad Gateway */
+            502: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "text/plain": string;
+                };
+            };
+            /** @description Insufficient Storage */
+            507: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "text/plain": string;
+                };
+            };
+        };
+    };
+    submitAgentSessionToolOutputs: {
+        parameters: {
+            query?: never;
+            header?: never;
+            path: {
+                session_id: string;
+            };
+            cookie?: never;
+        };
+        requestBody: {
+            content: {
+                "application/json": components["schemas"]["SubmitAgentSessionToolOutputsRequestBody"];
+            };
+        };
+        responses: {
+            /** @description Successful Response */
+            200: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": components["schemas"]["SubmitAgentSessionToolOutputsResponseBody"];
+                };
+            };
+            /** @description Bad Request */
+            400: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "text/plain": string;
+                };
+            };
+            /** @description Unauthorized */
+            401: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "text/plain": string;
+                };
+            };
+            /** @description Forbidden */
+            403: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "text/plain": string;
+                };
+            };
+            /** @description Not Found */
+            404: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "text/plain": string;
+                };
+            };
+            /** @description The request conflicts with the session's current state; the body's ``code`` field says how. */
+            409: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": components["schemas"]["AgentSessionConflictError"];
                 };
             };
             /** @description Validation Error */
@@ -11827,21 +12966,29 @@ export interface operations {
                     "application/json": components["schemas"]["HTTPValidationError"];
                 };
             };
+            /** @description Insufficient Storage */
+            507: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "text/plain": string;
+                };
+            };
         };
     };
-    summarize_endpoint_agents__agent_id__sessions__session_id__summary_post: {
+    agentSessionChat: {
         parameters: {
             query?: never;
             header?: never;
             path: {
-                agent_id: string;
                 session_id: string;
             };
             cookie?: never;
         };
         requestBody: {
             content: {
-                "application/json": components["schemas"]["_SummarizeRequest"];
+                "application/json": components["schemas"]["ChatRequestBody"];
             };
         };
         responses: {
@@ -11851,7 +12998,52 @@ export interface operations {
                     [name: string]: unknown;
                 };
                 content: {
-                    "application/json": components["schemas"]["_SummarizeResponse"];
+                    "application/json": unknown;
+                };
+            };
+            /** @description Bad Request */
+            400: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "text/plain": string;
+                };
+            };
+            /** @description Unauthorized */
+            401: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "text/plain": string;
+                };
+            };
+            /** @description Forbidden */
+            403: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "text/plain": string;
+                };
+            };
+            /** @description Not Found */
+            404: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "text/plain": string;
+                };
+            };
+            /** @description The request conflicts with the session's current state; the body's ``code`` field says how. */
+            409: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "application/json": components["schemas"]["AgentSessionConflictError"];
                 };
             };
             /** @description Validation Error */
@@ -11863,6 +13055,15 @@ export interface operations {
                     "application/json": components["schemas"]["HTTPValidationError"];
                 };
             };
+            /** @description Insufficient Storage */
+            507: {
+                headers: {
+                    [name: string]: unknown;
+                };
+                content: {
+                    "text/plain": string;
+                };
+            };
         };
     };
     version_arize_phoenix_version_get: {