improve agentcore docs (#20135)

Author: logan-markewich

docs/examples/memory/agentcore_memory_sample.ipynb

@@ -31,33 +31,38 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "%pip install llama-index llama-index-memory-bedrock-agentcore"
+    "%pip install llama-index llama-index-memory-bedrock-agentcore llama-index-llms-bedrock-converse"
    ]
   },
   {
    "cell_type": "markdown",
-   "id": "18cf5e60",
+   "id": "130ba50b",
    "metadata": {},
    "source": [
-    "Additionally, you can install the following requirements to follow along with the provided example. The yahoo finance tool is used in the example & the bedrock converse integration is used as the LLM for the FunctionAgent"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "id": "149b4dd8-5cd9-461e-b9cc-cc878034ea83",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "%pip install llama-index-tools-yahoo-finance llama-index-llms-bedrock-converse"
+    "### Bedrock AgentCore Setup Pre-Requisites\n",
+    "\n",
+    "1. AWS account with Bedrock AgentCore access\n",
+    "2. Configured AWS credentials (boto3)\n",
+    "3. Created memory resource in AWS Bedrock AgentCore\n",
+    "4. Required IAM permissions:\n",
+    "   1. bedrock-agentcore:CreateEvent\n",
+    "   2. bedrock-agentcore:ListEvents\n",
+    "   3. bedrock-agentcore:RetrieveMemories"
    ]
   },
   {
    "cell_type": "markdown",
    "id": "0df5a87e",
    "metadata": {},
    "source": [
-    "Initialize the AgentCore Memory context & AgentCore memory classes"
+    "## Setup\n",
+    "\n",
+    "Create an instance of AgentCoreMemoryContext to setup the memory resources that you will need for building an agent.\n",
+    "   1. Actor id → This is a required field and it is the identifier of the actor (could be an agent or the end-user).\n",
+    "   2. Memory id → This is a required field and it is the identifier of the memory store.\n",
+    "   3. Session id → This is a required field and it is the unique identifier of a particular conversation.\n",
+    "   4. Namespace → This is an optional field and it is used to determine how to extract long term memories. By default it will use “/” as the namespace.\n",
+    "   5. Memory strategy id → This is an optional field and it is the identifier for a memory strategy."
    ]
   },
   {
@@ -87,7 +92,9 @@
    "id": "7366bdc7",
    "metadata": {},
    "source": [
-    "Initialize the FunctionAgent or ReActAgent with any tool and LLM."
+    "Initialize the FunctionAgent or ReActAgent with any tool and LLM.\n",
+    "\n",
+    "Below we define some dummy tools to have the agent interact with."
    ]
   },
   {
@@ -99,13 +106,32 @@
    "source": [
     "from llama_index.llms.bedrock_converse import BedrockConverse\n",
     "from llama_index.core.agent.workflow import FunctionAgent\n",
-    "from llama_index.tools.yahoo_finance import YahooFinanceToolSpec\n",
+    "from llama_index.core.tools import FunctionTool\n",
+    "\n",
+    "\n",
+    "async def call_fn(name: str, message: str):\n",
+    "    \"\"\"Call the provided name.\n",
+    "    Args:\n",
+    "        name: str (Name of the person, will lookup their phone number)\n",
+    "        message: str (Message to send)\n",
+    "    \"\"\"\n",
+    "    print(f\"Called {name} and left a message.\")\n",
+    "\n",
+    "\n",
+    "async def email_fn(email: str, message: str):\n",
+    "    \"\"\"Email the provided name.\n",
+    "    Args:\n",
+    "        email: str (Email to send to)\n",
+    "        message: str (Body of the email)\n",
+    "    \"\"\"\n",
+    "    print(f\"Emailing {email} with the provided message.\")\n",
+    "\n",
     "\n",
     "llm = BedrockConverse(model=\"us.anthropic.claude-sonnet-4-20250514-v1:0\")\n",
     "\n",
-    "finance_tool_spec = YahooFinanceToolSpec()\n",
+    "\n",
     "agent = FunctionAgent(\n",
-    "    tools=finance_tool_spec.to_tool_list(),\n",
+    "    tools=[email_fn, call_fn],\n",
     "    llm=llm,\n",
     ")"
    ]
@@ -126,7 +152,8 @@
    "outputs": [],
    "source": [
     "response = await agent.run(\n",
-    "    \"What is the stock price for Amazon?\", memory=agentcore_memory\n",
+    "    \"Please email johan@coolcompany.com and tell him that his appointement has been moved to 3 PM January 10th, 2025.\",\n",
+    "    memory=agentcore_memory,\n",
     ")\n",
     "\n",
     "print(str(response))"
@@ -148,16 +175,26 @@
    "outputs": [],
    "source": [
     "response = await agent.run(\n",
-    "    \"What stock prices have I asked for?\", memory=agentcore_memory\n",
+    "    \"Who have I emailed recently?\", memory=agentcore_memory\n",
     ")\n",
     "\n",
     "print(str(response))"
    ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "fa55ebd9",
+   "metadata": {},
+   "source": [
+    "## References\n",
+    "\n",
+    "- [Bedrock AgentCore Memory Documentation](https://docs.aws.amazon.com/bedrock-agentcore/latest/devguide/memory-getting-started.html)"
+   ]
   }
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "Python 3 (ipykernel)",
+   "display_name": "llama_index",
    "language": "python",
    "name": "python3"
   },