Fix exercise inconsistencies across all three tracks

exercises/JavaScript/06-discover-connected-crimes.md

@@ -233,9 +233,22 @@ import { callRPT1Tool, callGroundingServiceTool, callSonarProSearchTool } from "
       const patternResult = await callSonarProSearchTool(patternQuery);
       intelligenceResults.push(`Similar Art Theft Patterns:\n${patternResult}`);
 
-      const intelligenceReport = `Intelligence Research Complete: ${intelligenceResults.join("\n\n")}
-      Summary: Conducted OSINT research on all suspects and identified similar crime patterns`;
+      const rawResults = `Intelligence Research Complete:\n\n${intelligenceResults.join("\n\n")}`;
 
+      const response = await this.orchestrationClient.chatCompletion({
+        messages: [
+          {
+            role: "system",
+            content: AGENT_CONFIGS.intelligenceResearcher.systemPrompt(state.suspect_names),
+          },
+          {
+            role: "user",
+            content: `Here are the web search results. Write a concise intelligence report:\n\n${rawResults}`,
+          },
+        ],
+      });
+
+      const intelligenceReport = response.getContent() || "No intelligence report could be generated.";
       console.log("✅ Intelligence research complete");
 
       return {

exercises/Python-LangGraph/06-discover-connected-crimes.md

@@ -148,11 +148,14 @@ def intelligence_researcher_node(state: AgentState) -> dict:
         pattern_result = call_sonar_pro_search(pattern_query)
         intelligence_results.append(f"Similar Art Theft Patterns:\n{pattern_result}")
 
-        intelligence_report = (
-            "Intelligence Research Complete:\n\n" + "\n\n".join(intelligence_results) +
-            f"\n\nSummary: Conducted OSINT research on all suspects and identified similar crime patterns"
-        )
+        raw_results = "Intelligence Research Complete:\n\n" + "\n\n".join(intelligence_results)
+
+        response = model.invoke([
+            SystemMessage(content=WEB_RESEARCHER_AGENT["prompt"]),
+            HumanMessage(content=f"Here are the web search results. Write a concise intelligence report:\n\n{raw_results}"),
+        ])
 
+        intelligence_report = response.content
         print("✅ Intelligence research complete")
 
         return {
@@ -304,7 +307,7 @@ print("="*50)
 print(result["intelligence_report"] or "(not set)")
 ```
 
-> 💡 **`server.py`** also constructs the initial state for `investigator_graph.ainvoke(...)`. Add `"intelligence_report": None` there too — find the dict and add the field.
+> 💡 **`server.py`** also constructs the initial state for `investigator_graph.invoke(...)`. Add `"intelligence_report": None` there too — find the dict and add the field.
 
 ---
 

exercises/Python-LangGraph/07-solve-the-crime.md

@@ -23,17 +23,20 @@ This is enough to make the system run, but the Lead Detective may stop short of
 👉 Update the `_lead_detective_prompt` function in `config/agents.py` with a more specific closing instruction:
 
 ```python
-def _lead_detective_prompt(appraisal_result: str, evidence_analysis: str, suspect_names: str) -> str:
+def _lead_detective_prompt(appraisal_result: str, evidence_analysis: str, intelligence_report: str, suspect_names: str) -> str:
     return (
         "You are the Lead Detective coordinating an art theft investigation. "
         "You have received the following information from your team:\n\n"
         f"1. INSURANCE APPRAISAL:\n{appraisal_result}\n\n"
-        f"2. EVIDENCE ANALYSIS:\n{evidence_analysis}\n\n"
-        f"3. SUSPECTS: {suspect_names}\n\n"
+        f"2. EVIDENCE ANALYSIS (Internal Documents):\n{evidence_analysis}\n\n"
+        f"3. INTELLIGENCE REPORT (Web Search):\n{intelligence_report}\n\n"
+        f"4. SUSPECTS: {suspect_names}\n\n"
         "Based on all the evidence and analysis, you MUST:\n"
         "- Name the most likely thief and explain the evidence supporting that conclusion\n"
+        "- Consider both internal evidence and external criminal patterns\n"
         "- Note any alibis or evidence that clears the other suspects\n"
         "- State the total insured value of the stolen goods\n"
+        "- Assess whether this is an isolated incident or part of a larger criminal network\n"
         "- Provide a comprehensive summary of the case."
     )
 ```

exercises/Python/04-building-multi-agent-system.md

@@ -26,13 +26,13 @@ Instead of defining agents and tasks directly in Python code, we'll move them to
 # agents.yaml
 appraiser_agent:
   role: >
-    Loss Appraiser
+    Stolen Goods Loss Appraiser
   goal: >
-    Predict the missing values of stolen items using the RPT-1 model via the call_rpt1 tool.
-    Use the payload data from inputs: {payload}
+    Predict the monetary value of stolen items ONLY by calling the call_rpt1 tool with payload {payload}.
+    Do NOT invent or estimate values yourself. If the tool call fails, report the failure.
   backstory: >
-    You are an expert insurance appraiser specializing in fine art valuation and theft assessment.
-  llm: sap/gpt-4o
+    You are an insurance appraiser who relies strictly on model predictions. You never guess values.
+  llm: sap/anthropic--claude-4.5-opus
 ```
 
 > 💡 **What's happening here?** We're taking the agent definition from your `basic_agent.py` file (the `role`, `goal`, `backstory`, and `llm` parameters) and moving them to this YAML file. The `>` symbol in YAML allows multi-line text.

exercises/Python/06-discover-connected-crimes.md

@@ -246,17 +246,6 @@ research_criminal_network:
         )
 ```
 
-👉 Also update the `solve_crime` task to use all three sources:
-
-```python
-    @task
-    def solve_crime(self) -> Task:
-        return Task(
-            config=self.tasks_config['solve_crime'],
-            context=[self.appraise_loss_task(), self.analyze_evidence_task(), self.research_criminal_network()]
-        )
-```
-
 > 💡 **Method Positioning Matters!**
 >
 > Your class should now have this order:

exercises/Python/07-solve-the-crime.md

@@ -57,25 +57,27 @@ Now you'll add the Lead Detective Agent and its task to your investigator crew.
     def solve_crime(self) -> Task:
         return Task(
             config=self.tasks_config['solve_crime'],
-            context=[self.appraise_loss_task(), self.analyze_evidence_task()]  # 👈 Lead detective uses results from other tasks
+            context=[self.appraise_loss_task(), self.analyze_evidence_task(), self.research_criminal_network()]  # 👈 Lead detective uses results from all three tasks
         )
 ```
 
-> 💡 **Where to place this code**: Add these methods inside the `InvestigatorCrew` class, after your `analyze_evidence_task()` method. The final order should be:
+> 💡 **Where to place this code**: Add these methods inside the `InvestigatorCrew` class, after your `research_criminal_network()` method from Exercise 06. The final order should be:
 >
 > 1. `appraiser_agent()` method
 > 2. `appraise_loss_task()` method
 > 3. `evidence_analyst_agent()` method
 > 4. `analyze_evidence_task()` method
-> 5. **👈 Add `lead_detective_agent()` here**
-> 6. **👈 Add `solve_crime()` here**
-> 7. `crew()` method (keep at the end)
+> 5. `intelligence_researcher_agent()` method (added in Exercise 06)
+> 6. `research_criminal_network()` method (added in Exercise 06)
+> 7. **👈 Add `lead_detective_agent()` here**
+> 8. **👈 Add `solve_crime()` here**
+> 9. `crew()` method (keep at the end)
 
 > 💡 **Understanding the `context` parameter:**
 >
-> - `context=[self.appraise_loss_task(), self.analyze_evidence_task()]` tells CrewAI that the `solve_crime` task depends on the other two tasks
-> - The Lead Detective will receive the output from both the Loss Appraiser and Evidence Analyst
-> - This enables the detective to combine financial predictions with evidence analysis to solve the crime
+> - `context=[self.appraise_loss_task(), self.analyze_evidence_task(), self.research_criminal_network()]` tells CrewAI that the `solve_crime` task depends on all three preceding tasks
+> - The Lead Detective will receive the output from the Loss Appraiser, Evidence Analyst, and Intelligence Researcher
+> - This enables the detective to combine financial predictions, evidence analysis, and criminal network intelligence to solve the crime
 
 ### Step 4: Verify Crew Configuration
 

project/JavaScript/solution/src/investigationWorkflow.ts

@@ -126,9 +126,22 @@ export class InvestigationWorkflow {
       const patternResult = await callSonarProSearchTool(patternQuery);
       intelligenceResults.push(`Similar Art Theft Patterns:\n${patternResult}`);
 
-      const intelligenceReport = `Intelligence Research Complete: ${intelligenceResults.join("\n\n")}
-      Summary: Conducted OSINT research on all suspects and identified similar crime patterns`;
+      const rawResults = `Intelligence Research Complete:\n\n${intelligenceResults.join("\n\n")}`;
 
+      const response = await this.orchestrationClient.chatCompletion({
+        messages: [
+          {
+            role: "system",
+            content: AGENT_CONFIGS.intelligenceResearcher.systemPrompt(state.suspect_names),
+          },
+          {
+            role: "user",
+            content: `Here are the web search results. Write a concise intelligence report:\n\n${rawResults}`,
+          },
+        ],
+      });
+
+      const intelligenceReport = response.getContent() || "No intelligence report could be generated.";
       console.log("✅ Intelligence research complete");
 
       return {

project/Python-LangGraph/solution/investigator_graph.py

@@ -173,11 +173,14 @@ def intelligence_researcher_node(state: AgentState) -> dict:
         pattern_result = call_sonar_pro_search(pattern_query)
         intelligence_results.append(f"Similar Art Theft Patterns:\n{pattern_result}")
 
-        intelligence_report = (
-            "Intelligence Research Complete:\n\n" + "\n\n".join(intelligence_results) +
-            f"\n\nSummary: Conducted OSINT research on all suspects and identified similar crime patterns"
-        )
+        raw_results = "Intelligence Research Complete:\n\n" + "\n\n".join(intelligence_results)
+
+        response = model.invoke([
+            SystemMessage(content=WEB_RESEARCHER_AGENT["prompt"]),
+            HumanMessage(content=f"Here are the web search results. Write a concise intelligence report:\n\n{raw_results}"),
+        ])
 
+        intelligence_report = response.content
         print("✅ Intelligence research complete")
 
         return {