From 662a432b9ef93cb1ead7d93f8b4759a7ae729a78 Mon Sep 17 00:00:00 2001
From: Ognjen Bostjancic <ognjen.bostjancic@gmail.com>
Date: Thu, 25 Jun 2026 12:36:43 +0200
Subject: [PATCH] feat(agent-monitoring): Add setUser step to Conversations
 onboarding

Add a guided 'Identify Users (optional)' step to the Conversations onboarding
flow, parallel to the existing 'Set Conversation ID' step. This teaches users
to call Sentry.setUser() (JS) or sentry_sdk.set_user() (Python) so the User
column in the Conversations view is populated.

Also mention setUser in the LLM_ONBOARDING_COPY_MARKDOWN prompt so AI coding
agents include the call when setting up agent monitoring.

Refs TET-2513
Co-Authored-By: Claude Sonnet 4 <noreply@anthropic.com>
---
 .../explore/conversations/onboarding.tsx      | 34 +++++++++++++++++++
 .../agents/llmOnboardingInstructions.tsx      |  1 +
 2 files changed, 35 insertions(+)

diff --git a/static/app/views/explore/conversations/onboarding.tsx b/static/app/views/explore/conversations/onboarding.tsx
index fd8ab8c8259b..3211db736500 100644
--- a/static/app/views/explore/conversations/onboarding.tsx
+++ b/static/app/views/explore/conversations/onboarding.tsx
@@ -273,6 +273,39 @@ Sentry.setConversationId("my-conversation-123");`,
   };
 }
 
+function getSetUserStep(isPython: boolean): OnboardingStep {
+  const content: ContentBlock[] = [
+    {
+      type: 'text',
+      text: t(
+        'Identify the user behind each conversation so the Conversations view can show who sent each message:'
+      ),
+    },
+    isPython
+      ? {
+          type: 'code' as const,
+          language: 'python',
+          code: `import sentry_sdk
+
+# Call this once per request / session, before any AI calls
+sentry_sdk.set_user({"id": "user_123", "email": "jane@example.com", "username": "jane"})`,
+        }
+      : {
+          type: 'code' as const,
+          language: 'javascript',
+          code: `import * as Sentry from "@sentry/node";
+
+// Call this once per request / session, before any AI calls
+Sentry.setUser({ id: "user_123", email: "jane@example.com", username: "jane" });`,
+        },
+  ];
+
+  return {
+    title: t('Identify Users (optional)'),
+    content,
+  };
+}
+
 export function ConversationOnboarding({onDismiss}: {onDismiss: () => void}) {
   const api = useApi();
   const {isSelfHosted, urlPrefix} = useLegacyStore(ConfigStore);
@@ -369,6 +402,7 @@ export function ConversationOnboarding({onDismiss}: {onDismiss: () => void}) {
     ...(agentMonitoringDocs.install?.(docParams) || []),
     ...(agentMonitoringDocs.configure?.(docParams) || []),
     getConversationIdStep(selectedIntegration, isPythonPlatform),
+    getSetUserStep(isPythonPlatform),
     ...(agentMonitoringDocs.verify?.(docParams) || []),
   ].filter(s => !s.collapsible);
 
diff --git a/static/app/views/insights/pages/agents/llmOnboardingInstructions.tsx b/static/app/views/insights/pages/agents/llmOnboardingInstructions.tsx
index f8a4623c9ff6..56cbdc43155e 100644
--- a/static/app/views/insights/pages/agents/llmOnboardingInstructions.tsx
+++ b/static/app/views/insights/pages/agents/llmOnboardingInstructions.tsx
@@ -70,6 +70,7 @@ export const LLM_ONBOARDING_COPY_MARKDOWN = `
 > The setup steps above contain the correct DSN and project-specific SDK configuration — complete them first.
 > Then follow the skill references below for instrumentation and agent naming.
 > If the app has multi-turn chats, set a conversation ID for each chat so Sentry can send the gen_ai.conversation.id attribute and show the session in Conversations.
+> Also call setUser (JS) / sentry_sdk.set_user (Python) once per request or session so conversations are attributed to users in the Conversations view.
 
 # Instrument Sentry AI Agent Monitoring