SAP-samples · davbizer93 · Apr 17, 2026 · Apr 17, 2026 · Jun 30, 2026 · Jul 8, 2026
diff --git a/README.md b/README.md
@@ -50,7 +50,10 @@ If you are unsure which path to choose, ask the instructor for guidance.
 - [Exercise 03 - Build your first agent tool](./exercises/Python/03-add-your-first-tool.md)
 - [Exercise 04 - Building a multi-agent system](./exercises/Python/04-building-multi-agent-system.md)
 - [Exercise 05 - Add the Grounding service](./exercises/Python/05-add-the-grounding-service.md)
-- [Exercise 06 - Use your AI Agents to solve the crime](./exercises/Python/06-solve-the-crime.md)
+- [Exercise 06 - Discover Connected Crimes with Web Search](./exercises/Python/06-discover-connected-crimes.md)
+- [Exercise 07 - Use your AI Agents to solve the crime](./exercises/Python/07-solve-the-crime.md)
+- [Exercise 08 - Deploy your agent to Cloud Foundry with A2A](./exercises/Python/08-deploy-agent-to-cf.md)
+- [Exercise 09 - Integrate your agent into SAP Joule](./exercises/Python/09-integrate-agent-into-joule.md)
 
 ### Python (LangGraph + LiteLLM)
 
@@ -60,8 +63,10 @@ If you are unsure which path to choose, ask the instructor for guidance.
 - [Exercise 03 - Build your first agent tool](./exercises/Python-LangGraph/03-add-your-first-tool.md)
 - [Exercise 04 - Building a multi-agent system](./exercises/Python-LangGraph/04-building-multi-agent-system.md)
 - [Exercise 05 - Add the Grounding service](./exercises/Python-LangGraph/05-add-the-grounding-service.md)
-- [Exercise 06 - Use your AI Agents to solve the crime](./exercises/Python-LangGraph/06-solve-the-crime.md)
-- [Exercise 07 - Deploy your agent to Cloud Foundry with A2A](./exercises/Python-LangGraph/07-deploy-agent-to-cf.md)
+- [Exercise 06 - Discover Connected Crimes with Web Search](./exercises/Python-LangGraph/06-discover-connected-crimes.md)
+- [Exercise 07 - Use your AI Agents to solve the crime](./exercises/Python-LangGraph/07-solve-the-crime.md)
+- [Exercise 08 - Deploy your agent to Cloud Foundry with A2A](./exercises/Python-LangGraph/08-deploy-agent-to-cf.md)
+- [Exercise 09 - Integrate your agent into SAP Joule](./exercises/Python-LangGraph/09-integrate-agent-into-joule.md)
 
 ### JavaScript (LangGraph + SAP Cloud SDK for AI)
 
@@ -71,8 +76,10 @@ If you are unsure which path to choose, ask the instructor for guidance.
 - [Exercise 03 - Build your first agent tool](./exercises/JavaScript/03-add-your-first-tool.md)
 - [Exercise 04 - Building a multi-agent system](./exercises/JavaScript/04-building-multi-agent-system.md)
 - [Exercise 05 - Add the Grounding service](./exercises/JavaScript/05-add-the-grounding-service.md)
-- [Exercise 06 - Use your AI Agents to solve the crime](./exercises/JavaScript/06-solve-the-crime.md)
-- [Exercise 07 - Deploy your agent to Cloud Foundry with A2A](./exercises/JavaScript/07-deploy-agent-to-cf-ts.md)
+- [Exercise 06 - Discover Connected Crimes with Web Search](./exercises/JavaScript/06-discover-connected-crimes.md)
+- [Exercise 07 - Use your AI Agents to solve the crime](./exercises/JavaScript/07-solve-the-crime.md)
+- [Exercise 08 - Deploy your agent to Cloud Foundry with A2A](./exercises/JavaScript/08-deploy-agent-to-cf-ts.md)
+- [Exercise 09 - Integrate your agent into SAP Joule](./exercises/JavaScript/09-integrate-agent-into-joule.md)
 
 The instructor will start you on the first exercise. Proceed to the next exercise once the instructor tells you to.
 

diff --git a/exercises/JavaScript/02-build-a-basic-agent.md b/exercises/JavaScript/02-build-a-basic-agent.md
@@ -399,7 +399,8 @@ In the following exercises, you will:
 4. 📌 [Add custom tools](03-add-your-first-tool.md) to your agents so they can access external data
 5. 📌 [Build a multi-agent workflow](04-building-multi-agent-system.md) with LangGraph
 6. 📌 [Integrate the Grounding Service](05-add-the-grounding-service.md) for evidence analysis
-7. 📌 [Solve the museum art theft mystery](06-solve-the-crime.md) using your fully-featured agent team
+7. 📌 [Discover Connected Crimes](06-discover-connected-crimes.md): Add web search with Perplexity sonar-pro
+8. 📌 [Solve the museum art theft mystery](07-solve-the-crime.md) using your fully-featured agent team
 
 ---
 

diff --git a/exercises/JavaScript/03-add-your-first-tool.md b/exercises/JavaScript/03-add-your-first-tool.md
@@ -496,7 +496,8 @@ In the following exercises, you will:
 4. ✅ Add custom tools to your agents so they can access external data (this exercise)
 5. 📌 [Build a multi-agent workflow](04-building-multi-agent-system.md) with LangGraph
 6. 📌 [Integrate the Grounding Service](05-add-the-grounding-service.md) for evidence analysis
-7. 📌 [Solve the museum art theft mystery](06-solve-the-crime.md) using your fully-featured agent team
+7. 📌 [Discover Connected Crimes](06-discover-connected-crimes.md): Add web search with Perplexity sonar-pro
+8. 📌 [Solve the museum art theft mystery](07-solve-the-crime.md) using your fully-featured agent team
 
 ---
 

diff --git a/exercises/JavaScript/04-building-multi-agent-system.md b/exercises/JavaScript/04-building-multi-agent-system.md
@@ -355,7 +355,7 @@ npx tsx src/main.ts
 
 > 💡 **Note:** The Evidence Analyst currently produces placeholder output because it doesn't have access to real evidence documents yet. You'll connect it to the Grounding Service in Exercise 05.
 
-> 💡 **Expected output:** The final report will say "Investigation completed but no conclusion was reached." — this is intentional. The Lead Detective node that produces `final_conclusion` hasn't been added yet; you'll implement it in Exercise 06.
+> 💡 **Expected output:** The final report will say "Investigation completed but no conclusion was reached." — this is intentional. The Lead Detective node that produces `final_conclusion` hasn't been added yet; you'll implement it in Exercise 07.
 
 ---
 
@@ -417,7 +417,7 @@ When you call `workflow.kickoff(inputs)`:
 
 - **Collaboration** — Agents can build upon each other's work
   - Sequential processing allows later nodes to use earlier results via shared state
-  - The `context` pattern (used in Exercise 06) enables explicit data sharing
+  - The `context` pattern (used in Exercise 07) enables explicit data sharing
 
 - **Maintainability** — Clear separation of concerns
   - Agent "personality" (goals, roles) lives in `agentConfigs.ts`
@@ -464,7 +464,8 @@ The Evidence Analyst can't access actual evidence yet. In Exercise 05, you'll in
 4. ✅ [Add custom tools](03-add-your-first-tool.md) (RPT-1 model integration)
 5. ✅ [Build a multi-agent workflow](04-building-multi-agent-system.md) (this exercise)
 6. 📌 [Add the Grounding Service](05-add-the-grounding-service.md): Give your Evidence Analyst access to real documents
-7. 📌 [Solve the crime](06-solve-the-crime.md): Add a Lead Detective to combine findings and crack the case
+7. 📌 [Discover Connected Crimes](06-discover-connected-crimes.md): Add web search with Perplexity sonar-pro
+8. 📌 [Solve the crime](07-solve-the-crime.md): Add a Lead Detective to combine findings and crack the case
 
 ---
 

diff --git a/exercises/JavaScript/05-add-the-grounding-service.md b/exercises/JavaScript/05-add-the-grounding-service.md
@@ -407,7 +407,8 @@ In the TypeScript approach, the orchestration pipeline handles the retrieval **a
 4. ✅ [Add custom tools](03-add-your-first-tool.md) to your agents
 5. ✅ [Build a multi-agent workflow](04-building-multi-agent-system.md)
 6. ✅ Integrate the Grounding Service (this exercise)
-7. 📌 [Solve the museum art theft mystery](06-solve-the-crime.md) using your fully-featured agent team
+7. 📌 [Add web search capabilities](06-discover-connected-crimes.md) to gather external intelligence
+8. 📌 [Solve the museum art theft mystery](07-solve-the-crime.md) using your fully-featured agent team
 
 ---
 
@@ -441,4 +442,4 @@ In the TypeScript approach, the orchestration pipeline handles the retrieval **a
 - [SAP Cloud SDK for AI — Orchestration Package](https://github.com/SAP/ai-sdk-js/tree/main/packages/orchestration)
 - [LangGraph.js Documentation](https://langchain-ai.github.io/langgraphjs/)
 
-[Next exercise](06-solve-the-crime.md)
+[Next exercise](06-discover-connected-crimes.md)