Spark
Sign In
Tool

Deploy LlamaIndex Agents to Bedrock AgentCore

Deploy LlamaIndex agents to Amazon Bedrock AgentCore Runtime with managed compute and memory.

Get this
Works with aws bedrockllama index
LlamaIndex
Own this? Claim it
82
Spark score
out of 100
Updated 2 days ago
Version 0.14.23
Models
claude
Add to Favorites

Why it matters

Deploy and manage LlamaIndex agents on Amazon Bedrock AgentCore Runtime. This asset provides a managed compute platform for your AI agents, enabling them to interact with sandboxed browser environments and execute code.

Outcomes

What it gets done

01

Deploy LlamaIndex agents to AWS Bedrock AgentCore Runtime.

02

Enable agents to perform sandboxed browser automation.

03

Integrate agents with code execution capabilities.

04

Manage agent memory and session context.

Install

Add it to your toolbox

Run in your project directory:

curl -fsSL https://spark.entire.vc/get/li-tool-tools-aws-bedrock-agentcore | bash

Overview

Amazon Bedrock AgentCore Runtime and Tools

A LlamaIndex runtime adapter that deploys agents to Amazon Bedrock AgentCore's managed compute, with streaming responses, persistent memory, and sandboxed browser tools. Use when deploying a working LlamaIndex agent to AWS's managed AgentCore platform instead of self-hosted infrastructure.

What it does

This module provides a runtime adapter and tools for deploying and extending LlamaIndex agents with Amazon Bedrock AgentCore, including managed compute via AgentCore Runtime, sandboxed browser automation, and code execution. The AgentCoreRuntime adapter wraps BedrockAgentCoreApp from the bedrock-agentcore SDK, exposing the required POST /invocations and GET /ping endpoints so any LlamaIndex agent can run on AgentCore's managed compute platform.

A single call, AgentCoreRuntime.serve(agent), starts a uvicorn server on port 8080 with sensible defaults. AgentCoreRuntime can also be instantiated directly with options: stream (SSE streaming, enabled by default), a required port for AgentCore deployment, and debug. Requests accept prompt, message, or input as the payload key, and can be sent non-streaming or as a streaming SSE connection. When streaming, the SSE stream emits typed events: agent_stream (token-by-token output with delta, response, and an optional thinking_delta), tool_call (before tool execution, with tool_name and tool_kwargs), tool_result (after tool execution, with tool_name and tool_output), done (the final response), and error (with a message).

The runtime also integrates with AgentCore Memory: an AgentCoreMemory instance configured with an AgentCoreMemoryContext (memory ID and actor ID) can be passed to AgentCoreRuntime.serve, and the session ID from the X-Amzn-Bedrock-AgentCore-Runtime-Session-Id header is automatically wired to it.

When to use - and when NOT to

Use it when you have a working LlamaIndex agent and want to deploy it to AWS's managed AgentCore compute platform rather than hosting your own server infrastructure, especially if you also want AgentCore's sandboxed browser automation or persistent memory across sessions. For testing without a full deployment, the adapter exposes runtime.app, a Starlette-based ASGI app usable directly with httpx.AsyncClient. Do not use it without valid AWS credentials and IAM permissions for bedrock-agentcore:* actions - the runtime is entirely dependent on AWS's AgentCore service being reachable and authorized.

Capabilities

AgentCoreRuntime.serve and the AgentCoreRuntime class deploy a LlamaIndex agent as an AgentCore-compatible service with streaming or non-streaming responses. AgentCoreMemory wires persistent, session-scoped memory into the runtime automatically from the request header. A Browser toolspec provides sandboxed browser automation tools, including navigate_browser to go to a URL and click_element to click an element by CSS selector.

How to install

pip install llama-index-tools-aws-bedrock-agentcore

Requires AWS credentials (via environment variables, an AWS CLI profile, or an IAM role), IAM permissions for bedrock-agentcore:* actions, and Python 3.9 or newer.

Who it's for

Developers who have built a LlamaIndex agent and want to deploy it to Amazon Bedrock AgentCore's managed runtime, with optional persistent memory and sandboxed browser automation, rather than operating their own agent-hosting infrastructure.

Source README

Amazon Bedrock AgentCore Runtime and Tools

This module provides a runtime adapter and tools for deploying and extending LlamaIndex agents with Amazon Bedrock AgentCore -- including managed compute via AgentCore Runtime, sandboxed browser automation, and code execution.

Prerequisites

  • AWS credentials configured via environment variables, AWS CLI profile, or IAM role
  • IAM permissions for bedrock-agentcore:* actions (see the AgentCore documentation for details)
  • Python 3.9+

Installation

(Optional) To run the examples below, first install:

pip install llama-index llama-index-llms-bedrock-converse

Install the main tools package:

pip install llama-index-tools-aws-bedrock-agentcore

Runtime

The AgentCoreRuntime adapter deploys any LlamaIndex agent to Amazon Bedrock AgentCore Runtime -- a managed compute platform for AI agents. It wraps BedrockAgentCoreApp from the bedrock-agentcore SDK, providing the required POST /invocations and GET /ping endpoints.

Quick Start

from llama_index.llms.bedrock_converse import BedrockConverse
from llama_index.core.agent.workflow import FunctionAgent
from llama_index.tools.aws_bedrock_agentcore import AgentCoreRuntime

llm = BedrockConverse(
    model="us.anthropic.claude-sonnet-4-6-v1",
    region_name="us-west-2",
)
agent = FunctionAgent(llm=llm, tools=[])

### One-liner -- starts uvicorn on port 8080
AgentCoreRuntime.serve(agent)

With Options

runtime = AgentCoreRuntime(
    agent=agent,
    stream=True,  # SSE streaming (default)
    port=8080,  # Required port for AgentCore deployment
    debug=False,
)
runtime.run()

With AgentCore Memory

from llama_index.memory.bedrock_agentcore import (
    AgentCoreMemory,
    AgentCoreMemoryContext,
)

memory = AgentCoreMemory(
    context=AgentCoreMemoryContext(
        memory_id="your-memory-id",
        actor_id="user-123",
    ),
    region_name="us-west-2",
)

### Session ID from the X-Amzn-Bedrock-AgentCore-Runtime-Session-Id header
### is automatically wired to memory
AgentCoreRuntime.serve(agent, memory=memory)

Sending Requests

### Non-streaming
curl -X POST http://localhost:8080/invocations \
  -H "Content-Type: application/json" \
  -d '{"prompt": "Hello, what can you do?"}'

### Streaming (SSE)
curl -N -X POST http://localhost:8080/invocations \
  -H "Content-Type: application/json" \
  -d '{"prompt": "Hello, what can you do?"}'

The adapter accepts prompt, message, or input as the payload key.

Streaming Event Types

When stream=True (default), the SSE stream emits these event types:

Event Fields Description
agent_stream delta, response, thinking_delta? Token-by-token LLM output
tool_call tool_name, tool_kwargs Before tool execution
tool_result tool_name, tool_output After tool execution
done response Final agent response
error message Error during streaming

Testing with ASGI

runtime = AgentCoreRuntime(agent=agent)
app = runtime.app  # BedrockAgentCoreApp (Starlette-based)
### Use with httpx.AsyncClient for testing

Toolspecs

Browser

The AgentCore Browser toolspec provides a set of tools for interacting with web browsers in a secure sandbox environment. It enables your LlamaIndex agents to navigate websites, extract content, click elements, and more.

Included tools:

  • navigate_browser: Navigate to a URL
  • click_element: Click on an element using CSS sele

FAQ

Common questions

Discussion

Questions & comments · 0

Sign In Sign in to leave a comment.