Add search_courses_tool to demonstrate catalog view + RAG pattern

The notebook mentioned that the agent could 'search the full catalog' but
didn't provide any tool to do so. Added a search_courses_tool that the agent
can use to retrieve detailed course information when needed, demonstrating
the pattern of using a high-level overview (catalog view) combined with
on-demand detailed retrieval (RAG).

Author: abrookins

python-recipes/context-engineering/notebooks/section-4-optimizations/05_crafting_data_for_llms.ipynb

@@ -365,12 +365,42 @@
         "\n",
         "catalog_view = redis_client.get(\"course_catalog_view\") or \"\"\n",
         "\n",
+        "# Define a tool for searching courses\n",
+        "from langchain_core.tools import tool\n",
+        "\n",
+        "@tool\n",
+        "async def search_courses_tool(query: str, limit: int = 5) -> str:\n",
+        "    \"\"\"Search for courses by semantic similarity to the query.\n",
+        "    \n",
+        "    Args:\n",
+        "        query: Natural language description of what courses to find\n",
+        "        limit: Maximum number of courses to return (default: 5)\n",
+        "    \n",
+        "    Returns:\n",
+        "        Formatted string with course details\n",
+        "    \"\"\"\n",
+        "    courses = await course_manager.search_courses(query=query, limit=limit)\n",
+        "    if not courses:\n",
+        "        return \"No courses found matching that query.\"\n",
+        "    \n",
+        "    result = []\n",
+        "    for course in courses:\n",
+        "        result.append(f\"\"\"Course: {course.course_code} - {course.title}\n",
+        "Department: {course.department}\n",
+        "Description: {course.description}\n",
+        "Credits: {course.credits} | Difficulty: {course.difficulty_level}\n",
+        "Format: {course.format}\"\"\")\n",
+        "    return \"\\n\\n\".join(result)\n",
+        "\n",
+        "# Bind the tool to the LLM\n",
+        "llm_with_tools = llm.bind_tools([search_courses_tool])\n",
+        "\n",
         "system_prompt = f\"\"\"You are a class scheduling agent for Redis University.\n",
         "\n",
         "{catalog_view}\n",
         "\n",
         "Use this overview to help students understand what's available.\n",
-        "For specific course details, you can search the full catalog.\n",
+        "For specific course details, use the search_courses_tool to find detailed information.\n",
         "\"\"\"\n",
         "\n",
         "user_query = \"What departments offer courses? I'm interested in computer science.\"\n",
@@ -380,11 +410,13 @@
         "    HumanMessage(content=user_query)\n",
         "]\n",
         "\n",
-        "response = llm.invoke(messages)\n",
+        "response = llm_with_tools.invoke(messages)\n",
         "\n",
         "print(f\"User: {user_query}\")\n",
         "print(f\"\\nAgent: {response.content}\")\n",
-        "print(\"\\n✅ Agent has high-level overview of entire catalog!\")"
+        "if response.tool_calls:\n",
+        "    print(f\"\\n🔧 Agent wants to use tools: {[tc['name'] for tc in response.tool_calls]}\")\n",
+        "print(\"\\n✅ Agent has high-level overview and can search for details!\")"
       ]
     },
     {