@agent-native/core 0.95.0 → 0.96.0

package/corpus/core/src/agent/production-agent.ts

@@ -1762,6 +1762,12 @@ export type AgentLoopFinalResponseGuardResult =
   | {
       retryMessage: string;
       fallbackMessage?: string;
+      /**
+       * Number of rejected text-only answers the model may correct before the
+       * fallback is emitted. Defaults to one and is capped to keep a broken
+       * guard/model combination from looping indefinitely.
+       */
+      maxRetries?: number;
     };
 
 export type AgentLoopFinalResponseGuard = (
@@ -3641,7 +3647,11 @@ export async function runAgentLoop(opts: {
           typeof guard === "string" ? guard : guard.retryMessage;
         const fallbackMessage =
           typeof guard === "string" ? guard : guard.fallbackMessage;
-        if (finalGuardRetries < 1) {
+        const maxGuardRetries =
+          typeof guard === "string"
+            ? 1
+            : Math.max(0, Math.min(3, Math.trunc(guard.maxRetries ?? 1)));
+        if (finalGuardRetries < maxGuardRetries) {
           finalGuardRetries += 1;
           send({ type: "clear" });
           messages.push({
@@ -3776,11 +3786,13 @@ export async function runAgentLoop(opts: {
         const result = sourceSweepGuard.message;
         send({
           type: "tool_start",
+          id: toolCall.id,
           tool: toolCall.name,
           input: toolCall.input as Record<string, string>,
         });
         send({
           type: "tool_done",
+          id: toolCall.id,
           tool: toolCall.name,
           input: toolCall.input as Record<string, unknown>,
           result,
@@ -3800,11 +3812,13 @@ export async function runAgentLoop(opts: {
         const result = sourceSweepDelegationGuard;
         send({
           type: "tool_start",
+          id: toolCall.id,
           tool: toolCall.name,
           input: toolCall.input as Record<string, string>,
         });
         send({
           type: "tool_done",
+          id: toolCall.id,
           tool: toolCall.name,
           input: toolCall.input as Record<string, unknown>,
           result,
@@ -3826,11 +3840,13 @@ export async function runAgentLoop(opts: {
         );
         send({
           type: "tool_start",
+          id: toolCall.id,
           tool: toolCall.name,
           input: toolCall.input as Record<string, string>,
         });
         send({
           type: "tool_done",
+          id: toolCall.id,
           tool: toolCall.name,
           input: toolCall.input as Record<string, unknown>,
           result,
@@ -3875,6 +3891,7 @@ export async function runAgentLoop(opts: {
         if (mustApprove) {
           send({
             type: "tool_start",
+            id: toolCall.id,
             tool: toolCall.name,
             input: toolCall.input as Record<string, string>,
           });
@@ -3913,6 +3930,7 @@ export async function runAgentLoop(opts: {
             `can run. The turn is paused; do not retry.`;
           send({
             type: "tool_done",
+            id: toolCall.id,
             tool: toolCall.name,
             input: toolCall.input as Record<string, unknown>,
             result,
@@ -3973,11 +3991,13 @@ export async function runAgentLoop(opts: {
             recordedResult;
           send({
             type: "tool_start",
+            id: toolCall.id,
             tool: toolCall.name,
             input: toolCall.input as Record<string, string>,
           });
           send({
             type: "tool_done",
+            id: toolCall.id,
             tool: toolCall.name,
             input: toolCall.input as Record<string, unknown>,
             result,
@@ -4027,11 +4047,13 @@ export async function runAgentLoop(opts: {
               ledgerResult;
             send({
               type: "tool_start",
+              id: toolCall.id,
               tool: toolCall.name,
               input: toolCall.input as Record<string, string>,
             });
             send({
               type: "tool_done",
+              id: toolCall.id,
               tool: toolCall.name,
               input: toolCall.input as Record<string, unknown>,
               result,
@@ -4056,11 +4078,13 @@ export async function runAgentLoop(opts: {
             `or split the request into smaller pieces.`;
           send({
             type: "tool_start",
+            id: toolCall.id,
             tool: toolCall.name,
             input: toolCall.input as Record<string, string>,
           });
           send({
             type: "tool_done",
+            id: toolCall.id,
             tool: toolCall.name,
             input: toolCall.input as Record<string, unknown>,
             result,
@@ -4107,6 +4131,7 @@ export async function runAgentLoop(opts: {
 
       send({
         type: "tool_start",
+        id: toolCall.id,
         tool: toolCall.name,
         input: toolCall.input as Record<string, string>,
       });
@@ -4122,6 +4147,7 @@ export async function runAgentLoop(opts: {
         );
         send({
           type: "tool_done",
+          id: toolCall.id,
           tool: toolCall.name,
           input: toolCall.input as Record<string, unknown>,
           result,
@@ -4153,6 +4179,7 @@ export async function runAgentLoop(opts: {
         );
         send({
           type: "tool_done",
+          id: toolCall.id,
           tool: toolCall.name,
           input: toolCall.input as Record<string, unknown>,
           result,
@@ -4184,6 +4211,7 @@ export async function runAgentLoop(opts: {
           `Previous result:\n${previousResult}`;
         send({
           type: "tool_done",
+          id: toolCall.id,
           tool: toolCall.name,
           input: toolCall.input as Record<string, unknown>,
           result,
@@ -4213,6 +4241,7 @@ export async function runAgentLoop(opts: {
         const result = planModeBlockedMessage(toolCall.name);
         send({
           type: "tool_done",
+          id: toolCall.id,
           tool: toolCall.name,
           input: toolCall.input as Record<string, unknown>,
           result,
@@ -4474,6 +4503,7 @@ export async function runAgentLoop(opts: {
 
       send({
         type: "tool_done",
+        id: toolCall.id,
         tool: toolCall.name,
         input: toolCall.input as Record<string, unknown>,
         result,
@@ -6140,21 +6170,82 @@ export function createProductionAgentHandler(
     // DIAGNOSTIC-ONLY: bracket stored-model resolution (getStoredModelForEngine
     // settings read).
     workerStep("model_start");
+    const storedModel =
+      requestModel == null && configuredModel == null
+        ? await getStoredModelForEngine(engine, { appId: options.appId })
+        : undefined;
     const modelCandidate =
-      requestModel ??
-      configuredModel ??
-      (await getStoredModelForEngine(engine, { appId: options.appId })) ??
-      engine.defaultModel;
+      requestModel ?? configuredModel ?? storedModel ?? engine.defaultModel;
     // DIAGNOSTIC-ONLY: stored-model resolution finished.
     workerStep("model_done");
     const model = normalizeModelForEngine(engine, modelCandidate);
+    let effectiveModel = model;
+    let modelSelectionSource:
+      | "request"
+      | "configured"
+      | "stored"
+      | "default"
+      | "experiment" =
+      requestModel != null
+        ? "request"
+        : configuredModel != null
+          ? "configured"
+          : storedModel != null
+            ? "stored"
+            : "default";
+    let experimentAssignments: Array<{
+      experimentId: string;
+      variantId: string;
+    }> = [];
+
+    // Database-backed experiments remain available to app operators. A
+    // platform-wide hosted rollout is layered in only when no explicit or
+    // persisted model selection exists, and only on the cross-provider Builder
+    // engine. This preserves the model picker's user intent.
+    try {
+      if (ownerEmail) {
+        const { resolveActiveExperimentConfig } =
+          await import("../observability/experiments.js");
+        const expConfig = await resolveActiveExperimentConfig(ownerEmail);
+        if (expConfig) {
+          experimentAssignments = [...expConfig.assignments];
+          if (typeof expConfig.configs.model === "string") {
+            effectiveModel = normalizeModelForEngine(
+              engine,
+              expConfig.configs.model,
+            );
+            modelSelectionSource = "experiment";
+          }
+        }
+
+        if (modelSelectionSource === "default") {
+          const { resolveHostedDefaultModelExperiment } =
+            await import("../observability/hosted-model-experiment.js");
+          const hostedExperiment = resolveHostedDefaultModelExperiment({
+            userId: ownerEmail,
+            engineName: engine.name,
+            isDefaultModelSelection: true,
+            supportedModels: engine.supportedModels,
+          });
+          if (hostedExperiment) {
+            effectiveModel = hostedExperiment.model;
+            experimentAssignments.push(hostedExperiment.assignment);
+            modelSelectionSource = "experiment";
+          }
+        }
+      }
+    } catch {
+      // Experiments are best-effort. Model resolution must keep working if the
+      // observability tables are unavailable during startup or migration.
+    }
+
     const reasoningEffort = resolveAgentRequestReasoningEffort({
-      model,
+      model: effectiveModel,
       requestEffort,
       configuredEffort: options.reasoningEffort,
     });
 
-    options.onEngineResolved?.(engine, model);
+    options.onEngineResolved?.(engine, effectiveModel);
 
     // One-line per-turn resolution log so it's obvious in dev which engine
     // is actually handling the request. `requestEngine` is what the client
@@ -6162,7 +6253,7 @@ export function createProductionAgentHandler(
     // Divergence between them is the usual cause of "status says builder but
     // no [builder-engine] log lines appear" confusion.
     console.log(
-      `[agent-chat] resolved engine=${engine.name} model=${model} requestEngine=${requestEngine ?? "(none)"}`,
+      `[agent-chat] resolved engine=${engine.name} model=${effectiveModel} requestEngine=${requestEngine ?? "(none)"} modelSource=${modelSelectionSource}`,
     );
 
     if (
@@ -7467,25 +7558,6 @@ export function createProductionAgentHandler(
           }
         }
 
-        // Apply experiment variant overrides (A/B testing)
-        let effectiveModel = model;
-        try {
-          const { resolveActiveExperimentConfig } =
-            await import("../observability/experiments.js");
-          if (!ownerEmail) {
-            // Without an authenticated owner we can't resolve user-scoped experiments.
-            throw new Error("no authenticated user");
-          }
-          const expConfig = await resolveActiveExperimentConfig(ownerEmail);
-          if (expConfig) {
-            if (typeof expConfig.configs.model === "string") {
-              effectiveModel = expConfig.configs.model;
-            }
-          }
-        } catch {
-          // Experiments module unavailable — use default model
-        }
-
         // TODO(processor-seam): thread `processors` from ProductionAgentOptions
         // through to runAgentLoop here once the handler exposes a way to
         // configure them (e.g. a `processors` field on ProductionAgentOptions
@@ -7572,6 +7644,14 @@ export function createProductionAgentHandler(
               threadId: threadId ?? null,
               userId: ownerEmail,
               config: obsConfig,
+              metadata: {
+                modelSelectionSource,
+                ...(experimentAssignments.length > 0
+                  ? { experimentAssignments }
+                  : {}),
+              },
+              experimentAssignments,
+              modelSelectionSource,
               classifyError: () => {
                 if (
                   agentLoopOpts.signal.aborted &&

package/corpus/core/src/agent/thread-data-builder.ts

@@ -1,5 +1,6 @@
 import type { ActionChatUIConfig } from "../action-ui.js";
 import {
+  isCredentialGapCodeAgentEvent,
   normalizeCodeAgentTranscript,
   type CodeAgentTranscriptEvent as CoreCodeAgentTranscriptEvent,
   type NormalizedCodeAgentStatusEvent,
@@ -152,9 +153,9 @@ export function buildAssistantMessage(
 
     if (event.type === "tool_start") {
       toolCallCounter += 1;
-      const toolCallId = runId
-        ? `${runId}:tc_${toolCallCounter}`
-        : `tc_${toolCallCounter}`;
+      const toolCallId =
+        event.id?.trim() ||
+        (runId ? `${runId}:tc_${toolCallCounter}` : `tc_${toolCallCounter}`);
       const args = (event.input ?? {}) as Record<string, string>;
       content.push({
         type: "tool-call",
@@ -167,23 +168,47 @@ export function buildAssistantMessage(
     }
 
     if (event.type === "tool_done") {
-      for (let i = content.length - 1; i >= 0; i--) {
-        const part = content[i];
-        if (
-          part.type === "tool-call" &&
-          part.toolName === event.tool &&
-          part.result === undefined
-        ) {
-          part.result = event.result ?? "";
-          if (event.isError !== undefined) part.isError = event.isError;
-          if (event.completedSideEffect !== undefined) {
-            part.completedSideEffect = event.completedSideEffect;
+      const eventToolCallId = event.id?.trim();
+      let matchingIndex = -1;
+
+      if (eventToolCallId) {
+        for (let i = content.length - 1; i >= 0; i--) {
+          const part = content[i];
+          if (
+            part.type === "tool-call" &&
+            part.toolCallId === eventToolCallId &&
+            part.result === undefined
+          ) {
+            matchingIndex = i;
+            break;
           }
-          if (event.mcpApp) part.mcpApp = event.mcpApp;
-          if (event.chatUI) part.chatUI = event.chatUI;
-          break;
         }
       }
+
+      if (matchingIndex === -1) {
+        for (let i = content.length - 1; i >= 0; i--) {
+          const part = content[i];
+          if (
+            part.type === "tool-call" &&
+            part.toolName === event.tool &&
+            part.result === undefined
+          ) {
+            matchingIndex = i;
+            break;
+          }
+        }
+      }
+
+      const part = content[matchingIndex];
+      if (part?.type === "tool-call") {
+        part.result = event.result ?? "";
+        if (event.isError !== undefined) part.isError = event.isError;
+        if (event.completedSideEffect !== undefined) {
+          part.completedSideEffect = event.completedSideEffect;
+        }
+        if (event.mcpApp) part.mcpApp = event.mcpApp;
+        if (event.chatUI) part.chatUI = event.chatUI;
+      }
       continue;
     }
 
@@ -640,6 +665,7 @@ export interface CodeAgentThreadTranscriptEvent {
   metadata?: Record<string, unknown>;
   artifactPath?: string;
   artifactUrl?: string;
+  signal?: CoreCodeAgentTranscriptEvent["signal"];
 }
 
 export interface BuildRepositoryFromCodeAgentTranscriptOptions {
@@ -1070,6 +1096,7 @@ function toCoreCodeAgentTranscriptEvent(
       ...(event.artifactPath ? { artifactPath: event.artifactPath } : {}),
       ...(event.artifactUrl ? { artifactUrl: event.artifactUrl } : {}),
     },
+    ...(event.signal ? { signal: event.signal } : {}),
   };
 }
 
@@ -1113,6 +1140,9 @@ function toolContentPartForCodeAgentTranscriptItem(
       ? { result: previewCodeAgentTranscriptValue(item.result) ?? "" }
       : {}),
     ...(item.structuredMeta ? { structuredMeta: item.structuredMeta } : {}),
+    ...(item.pendingApprovalKey
+      ? { approval: { approvalKey: item.pendingApprovalKey } }
+      : {}),
   };
 }
 
@@ -1120,7 +1150,7 @@ function statusTextForCodeAgentTranscriptItem(
   item: NormalizedCodeAgentStatusEvent,
   options: BuildRepositoryFromCodeAgentTranscriptOptions,
 ): string | null {
-  if (options.hideCredentialMessages && isCredentialCodeAgentText(item.text)) {
+  if (options.hideCredentialMessages && isCredentialGapCodeAgentEvent(item)) {
     return null;
   }
   if (item.statusKind === "artifact") {
@@ -1195,10 +1225,6 @@ function stringRecordValue(
   return typeof value === "string" && value.trim() ? value.trim() : undefined;
 }
 
-function isCredentialCodeAgentText(value: string): boolean {
-  return /No LLM provider key was found|Missing credentials/i.test(value);
-}
-
 export function upsertUserMessage(repo: any, userMsg: UserMessage): any {
   const nextRepo = normalizeThreadRepository(repo);
 

package/corpus/core/src/chat-threads/store.ts

@@ -639,6 +639,14 @@ export interface ListThreadsOptions {
   /** When true, returns only threads with no scope (general chats). */
   unscopedOnly?: boolean;
   orgId?: string | null;
+  /**
+   * Include archived threads in the results. Defaults to false: archived
+   * threads (`archived_at` set via `setThreadArchived`) are hidden from the
+   * ordinary chat list/search so archiving actually removes a thread from
+   * view. Pass true for surfaces that explicitly need to see archived chats
+   * (e.g. an "Archived" filter or restoring one via `setThreadArchived`).
+   */
+  includeArchived?: boolean;
 }
 
 function chatThreadAccessSql(
@@ -686,6 +694,9 @@ export async function listThreads(
   );
   const filters: string[] = [access.sql, `message_count > 0`];
   const args: (string | number)[] = [...access.args];
+  if (!opts.includeArchived) {
+    filters.push(`archived_at IS NULL`);
+  }
   if (opts.scope) {
     filters.push(`scope_type = ? AND scope_id = ?`);
     args.push(opts.scope.type, opts.scope.id);
@@ -710,7 +721,12 @@ export async function searchThreads(
   ownerEmail: string,
   query: string,
   limit = 50,
-  options: { scope?: { type: string; id: string }; orgId?: string | null } = {},
+  options: {
+    scope?: { type: string; id: string };
+    orgId?: string | null;
+    /** See `ListThreadsOptions.includeArchived` — defaults to false. */
+    includeArchived?: boolean;
+  } = {},
 ): Promise<ChatThreadSummary[]> {
   await ensureTable();
   const client = getDbExec();
@@ -728,6 +744,9 @@ export async function searchThreads(
     `(title LIKE ? ESCAPE '!' OR preview LIKE ? ESCAPE '!' OR thread_data LIKE ? ESCAPE '!')`,
   ];
   const args: (string | number)[] = [...access.args, pattern, pattern, pattern];
+  if (!options.includeArchived) {
+    filters.push(`archived_at IS NULL`);
+  }
   if (options.scope) {
     filters.push(`scope_type = ? AND scope_id = ?`);
     args.push(options.scope.type, options.scope.id);

package/corpus/core/src/cli/code-agent-executor.ts

@@ -215,6 +215,7 @@ export async function executeCodeAgentRun(
       runId: existing.id,
       kind: "status",
       message,
+      signal: "credential-gap",
       metadata: { status: "paused", phase: "missing-credentials" },
     });
     return updateCodeAgentRunRecord(existing.id, {

package/corpus/core/src/cli/code-agent-runs.ts

@@ -77,6 +77,17 @@ export type CodeAgentTranscriptEventKind =
   | "artifact"
   | "status";
 
+/**
+ * Structured, machine-checkable marker for transcript events that need
+ * special handling in the UI beyond free-text matching. `"credential-gap"`
+ * marks the status event the executor appends when no LLM provider key (or
+ * Codex CLI login) is available; consumers should prefer this field over
+ * regex-matching `message` (see `isCredentialGapCodeAgentEvent` in
+ * `../code-agents/transcript-normalizer.js`). Optional so older, already
+ * persisted JSONL transcripts without the field keep parsing unchanged.
+ */
+export type CodeAgentTranscriptEventSignal = "credential-gap";
+
 export interface CodeAgentTranscriptEvent {
   schemaVersion: 1;
   id: string;
@@ -85,6 +96,7 @@ export interface CodeAgentTranscriptEvent {
   message: string;
   createdAt: string;
   metadata?: Record<string, unknown>;
+  signal?: CodeAgentTranscriptEventSignal;
 }
 
 export interface CreateCodeAgentRunInput {
@@ -109,6 +121,7 @@ export interface AppendCodeAgentTranscriptEventInput {
   message: string;
   createdAt?: string;
   metadata?: Record<string, unknown>;
+  signal?: CodeAgentTranscriptEventSignal;
 }
 
 export interface QueueCodeAgentFollowUpInput {
@@ -294,6 +307,7 @@ export function appendCodeAgentTranscriptEvent(
     message: input.message,
     createdAt,
     metadata: input.metadata,
+    ...(input.signal ? { signal: input.signal } : {}),
   };
 
   fs.mkdirSync(codeAgentTranscriptsDir(), { recursive: true });

package/corpus/core/src/client/AgentPanel.tsx

@@ -1458,32 +1458,39 @@ function AgentPanelInner({
         activeTab && (activeTabMessageCount > 0 || activeTab.status !== "idle");
 
       return (
-        <div className="pointer-events-none absolute inset-x-0 top-3 z-[60] flex justify-end px-3 sm:top-4 sm:px-4">
-          <div className="pointer-events-auto flex items-center gap-1">
-            {canShareActiveTab ? (
-              <ShareButton
-                resourceType="chat_thread"
-                resourceId={activeTab.id}
-                resourceTitle={activeTab.label || "Chat"}
-                shareUrl={getChatThreadShareUrl(activeTab.id)}
-                trigger="icon"
-                triggerClassName="h-8 w-8 border border-border bg-background/95 shadow-sm backdrop-blur hover:bg-accent"
-              />
-            ) : null}
-            <button
-              type="button"
-              data-agent-page-new-chat=""
-              aria-label={t("agentPanel.newChat")}
-              onClick={() => {
-                addTab();
-              }}
-              className="inline-flex h-8 items-center gap-1.5 rounded-md border border-border bg-background/95 px-2.5 text-xs font-medium text-foreground shadow-sm backdrop-blur transition-colors hover:bg-accent focus-visible:outline-none focus-visible:ring-2 focus-visible:ring-ring"
-            >
-              <IconPlus size={14} />
-              <span>{t("agentPanel.newChat")}</span>
-            </button>
+        <>
+          <div
+            aria-hidden="true"
+            data-agent-page-chat-fade=""
+            className="pointer-events-none absolute inset-x-0 top-0 z-50 h-16 bg-gradient-to-b from-background via-background/90 to-transparent opacity-0 transition-opacity duration-150"
+          />
+          <div className="pointer-events-none absolute inset-x-0 top-3 z-[60] flex justify-end px-3 sm:top-4 sm:px-4">
+            <div className="pointer-events-auto flex items-center gap-1">
+              {canShareActiveTab ? (
+                <ShareButton
+                  resourceType="chat_thread"
+                  resourceId={activeTab.id}
+                  resourceTitle={activeTab.label || "Chat"}
+                  shareUrl={getChatThreadShareUrl(activeTab.id)}
+                  trigger="icon"
+                  triggerClassName="h-8 w-8 border border-border bg-background/95 shadow-sm backdrop-blur hover:bg-accent"
+                />
+              ) : null}
+              <button
+                type="button"
+                data-agent-page-new-chat=""
+                aria-label={t("agentPanel.newChat")}
+                onClick={() => {
+                  addTab();
+                }}
+                className="inline-flex h-8 items-center gap-1.5 rounded-md border border-border bg-background/95 px-2.5 text-xs font-medium text-foreground shadow-sm backdrop-blur transition-colors hover:bg-accent focus-visible:outline-none focus-visible:ring-2 focus-visible:ring-ring"
+              >
+                <IconPlus size={14} />
+                <span>{t("agentPanel.newChat")}</span>
+              </button>
+            </div>
           </div>
-        </div>
+        </>
       );
     },
     [getChatThreadShareUrl, t],