Skip to main content

An SDK for using Bedrock AgentCore

Project description

Bedrock AgentCore SDK

Deploy your local AI agent to Bedrock AgentCore with zero infrastructure

GitHub commit activity GitHub open issues GitHub open pull requests License PyPI version Python versions

DocumentationSamplesDiscordBoto3 Python SDKRuntime Python SDK

Overview

Amazon Bedrock AgentCore enables you to deploy and operate highly effective agents securely, at scale using any framework and model. With Amazon Bedrock AgentCore, developers can accelerate AI agents into production with the scale, reliability, and security, critical to real-world deployment. AgentCore provides tools and capabilities to make agents more effective and capable, purpose-built infrastructure to securely scale agents, and controls to operate trustworthy agents. Amazon Bedrock AgentCore services are composable and work with popular open-source frameworks and any model, so you don’t have to choose between open-source flexibility and enterprise-grade security and reliability.

🚀 From Local Development to Bedrock AgentCore

from bedrock_agentcore import BedrockAgentCoreApp
app = BedrockAgentCoreApp()

from strands import Agent # or bring your agent.

@app.entrypoint
async def handler(request):
    prompt = request.get("prompt")

    agent = Agent()

    async for event in agent.stream_async(prompt):
        yield (event)

app.run()

What you get with Bedrock AgentCore:

  • Keep your agent logic - Works with Strands, LangGraph, CrewAI, Autogen, or custom frameworks
  • Zero infrastructure management - No servers, containers, or scaling concerns
  • Enterprise-grade platform - Built-in auth, memory, observability, security
  • Production-ready deployment - Reliable, scalable, compliant hosting

Amazon Bedrock AgentCore services

  • 🚀 Runtime - Secure and session isolated compute: Get started
  • 🧠 Memory - Persistent knowledge across sessions: Get started
  • 🔗 Gateway - Transform APIs into MCP tools: Get started
  • 💻 Code Interpreter - Secure sandboxed execution: Get started
  • 🌐 Browser - Cloud-based web automation: Get started
  • 📊 Observability - OpenTelemetry tracing: Get started
  • 🔐 Identity - AWS & third-party auth: Get started

AG-UI Protocol Support

Deploy agents using the AG-UI protocol with zero boilerplate. A single entrypoint handler is served over both SSE (POST /invocations) and WebSocket (/ws).

from bedrock_agentcore.runtime import serve_ag_ui

# Framework agent with a .run() method — one line
serve_ag_ui(agui_agent)

Or write a custom agent with the decorator form:

from bedrock_agentcore.runtime import AGUIApp
from ag_ui.core import RunAgentInput, RunStartedEvent, RunFinishedEvent

app = AGUIApp()

@app.entrypoint
async def my_agent(input_data: RunAgentInput):
    yield RunStartedEvent(thread_id=input_data.thread_id, run_id=input_data.run_id)
    # ... your agent logic, yield AG-UI events ...
    yield RunFinishedEvent(thread_id=input_data.thread_id, run_id=input_data.run_id)

app.run()

Install with: pip install "bedrock-agentcore[ag-ui]"

See the AG-UI protocol contract for full details.

🏗️ Deployment

AgentCore CLI (Recommended)

The fastest way to deploy your agent to Bedrock AgentCore. Install via npm:

npm i -g @aws/agentcore

Then create and deploy your agent:

agentcore create --name MyAgent --defaults
cd MyAgent
agentcore deploy

The CLI handles packaging, infrastructure provisioning, and deployment automatically. See the AgentCore CLI documentation for full details.

AWS CDK (Advanced)

The AgentCore CLI generates AWS CDK under the hood. For full infrastructure-as-code control, customize the generated CDK project directly or refer to the L1 CDK documentation.

A2A Protocol Support

Serve your agent using the A2A (Agent-to-Agent) protocol on Bedrock AgentCore Runtime. Works with any framework that provides an a2a-sdk AgentExecutor (Strands, LangGraph, Google ADK, or custom).

pip install "bedrock-agentcore[a2a]"
from strands import Agent
from strands.a2a import StrandsA2AExecutor
from bedrock_agentcore.runtime import serve_a2a

agent = Agent(model="us.anthropic.claude-sonnet-4-20250514", system_prompt="You are a helpful assistant.")
serve_a2a(StrandsA2AExecutor(agent))

See A2A Protocol Examples for LangGraph, Google ADK, and advanced usage.

📝 License & Contributing

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

bedrock_agentcore-1.8.0.tar.gz (579.8 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

bedrock_agentcore-1.8.0-py3-none-any.whl (206.1 kB view details)

Uploaded Python 3

File details

Details for the file bedrock_agentcore-1.8.0.tar.gz.

File metadata

  • Download URL: bedrock_agentcore-1.8.0.tar.gz
  • Upload date:
  • Size: 579.8 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.12

File hashes

Hashes for bedrock_agentcore-1.8.0.tar.gz
Algorithm Hash digest
SHA256 99fb4fafed0e832bcfe0c588572ecc5cdf296692052f1f0607dc656510aa24d0
MD5 d0842a34aa0792050f0d184ece9aa18c
BLAKE2b-256 e556992ff28a1aaf674f525ea82c0f1e2d940bdf91c08322d8c258a2b1469dbd

See more details on using hashes here.

Provenance

The following attestation bundles were made for bedrock_agentcore-1.8.0.tar.gz:

Publisher: release-publish.yml on aws/bedrock-agentcore-sdk-python

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file bedrock_agentcore-1.8.0-py3-none-any.whl.

File metadata

File hashes

Hashes for bedrock_agentcore-1.8.0-py3-none-any.whl
Algorithm Hash digest
SHA256 0b0146cbc6a31dc5d17efafc62a5b87bec98b5bcd7cdf8ec26507eb6448de02e
MD5 5e4c2f9fdd9a03762eb894f5b1038be1
BLAKE2b-256 24281a47ee1cf275a9a205cb646db12ff42d0c59869112ef636727773b87ce6b

See more details on using hashes here.

Provenance

The following attestation bundles were made for bedrock_agentcore-1.8.0-py3-none-any.whl:

Publisher: release-publish.yml on aws/bedrock-agentcore-sdk-python

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page