Skip to content

Open-source SEO, GEO, and marketing skills for AI agents.

fix(chat): scope working indicator to the current turn

Yuting Zhong80adc2178646
diff --git a/src/components/live-transcript.test.tsx b/src/components/live-transcript.test.tsx
index 707c73b3..05ddf3d9 100644
--- a/src/components/live-transcript.test.tsx
+++ b/src/components/live-transcript.test.tsx
@@ -97,10 +97,10 @@ describe("LiveTranscript empty state", () => {
screen.queryByText(/no messages yet/i),
).not.toBeInTheDocument();
// WorkingIndicator renders the agent display name as the accent
- // header. With no events yet, mood is "waiting" and the headline
- // is "<agent> is starting".
+ // header and a verb-only headline next to it. With no events yet,
+ // mood is "waiting" and the headline reads "Starting".
expect(screen.getByText("CMO")).toBeInTheDocument();
- expect(screen.getByText(/is starting/i)).toBeInTheDocument();
+ expect(screen.getByText(/^Starting$/)).toBeInTheDocument();
});
});
diff --git a/src/components/live-transcript.tsx b/src/components/live-transcript.tsx
index 54774d5f..a135f63c 100644
--- a/src/components/live-transcript.tsx
+++ b/src/components/live-transcript.tsx
@@ -1085,6 +1085,11 @@ function LiveWorkingIndicator({
lifecyclePhase: lifecyclePhase ?? null,
pendingTools: pendingTools ?? [],
hasPendingAssistant: hasPendingAssistant ?? false,
+ // The indicator must only reflect the current turn — without this
+ // filter, the trajectory chips show every tool call from the chat's
+ // entire history (e.g. a fresh "hi" lit up runScript / runScript /
+ // listConnectedAccounts from prior audit turns).
+ turnStartedAt,
});
// Anchor elapsed to whichever is later: the turn-start wallclock the
@@ -1141,21 +1146,34 @@ function deriveWorkingView(input: {
lifecyclePhase: string | null;
pendingTools: ToolEntry[];
hasPendingAssistant: boolean;
+ /**
+ * Wallclock when the current turn started. We use it to scope the
+ * trajectory chips (and the in-flight tool detection) to the active
+ * turn — events from earlier turns in the same thread don't belong
+ * in "what is the agent doing right now?".
+ */
+ turnStartedAt: number | null;
}): WorkingView {
- const { agentDisplayName, events, lifecyclePhase, pendingTools, hasPendingAssistant } = input;
- const lastEvent = events.length > 0 ? events[events.length - 1] : null;
+ const { agentDisplayName, events, lifecyclePhase, pendingTools, hasPendingAssistant, turnStartedAt } = input;
+ // Only consider events from the current turn. Anything older is part
+ // of the persistent transcript above this card, not the live status.
+ const turnEvents =
+ turnStartedAt != null
+ ? events.filter((e) => e.ts >= turnStartedAt)
+ : events;
+ const lastEvent = turnEvents.length > 0 ? turnEvents[turnEvents.length - 1] : null;
const pendingInFlightTool = pendingTools.find((t) => !t.done) ?? null;
const inFlightCommittedToolCall: Extract<
TranscriptEvent,
{ kind: "tool_call" }
> | null = (() => {
if (!lastEvent || lastEvent.kind !== "tool_call") return null;
- const matched = events.some(
+ const matched = turnEvents.some(
(e) => e.kind === "tool_result" && e.tool_call_id === lastEvent.tool_call_id,
);
return matched ? null : lastEvent;
})();
- const phases = buildPhases(events, pendingTools);
+ const phases = buildPhases(turnEvents, pendingTools);
if (pendingInFlightTool) {
return {
@@ -1168,7 +1186,7 @@ function deriveWorkingView(input: {
if (hasPendingAssistant) {
return {
headline: "Writing the response",
- subtitle: subtitleForLastTool(pendingTools, events),
+ subtitle: subtitleForLastTool(pendingTools, turnEvents),
phases,
mood: "writing",
};
@@ -1178,7 +1196,9 @@ function deriveWorkingView(input: {
? humanLifecyclePhase(lifecyclePhase)
: null;
return {
- headline: `${agentDisplayName} is starting`,
+ // Agent name is rendered separately as the accent prefix in the
+ // indicator, so headlines stay tight + verb-only here.
+ headline: "Starting",
subtitle: lifecycleSummary ?? "Delivering the brief to OpenClaw",
phases,
mood: "waiting",
@@ -1194,7 +1214,7 @@ function deriveWorkingView(input: {
}
if (lastEvent.kind === "tool_result") {
return {
- headline: `${agentDisplayName} is thinking`,
+ headline: "Thinking",
subtitle: lastEvent.ok
? `${formatToolName(lastEvent.name)} ✓ — picking next step`
: `${formatToolName(lastEvent.name)} failed — retrying`,