Skip to content

strands-agentcore-tools

strands-agentcore-tools provides agentic tools that enable AI agents to autonomously deploy, manage, and monitor themselves on AWS Bedrock AgentCore.

Installation

pip install strands-agentcore-tools

Usage

from strands import Agent
from strands_agentcore_tools import configure, launch, invoke, status, logs

agent = Agent(
    tools=[configure, launch, invoke, status, logs],
    system_prompt="You can deploy yourself to AWS AgentCore."
)

# Agent deploys itself
agent("""
Deploy yourself to AWS:
1. Configure deployment
2. Launch to production
3. Check status
4. Show logs
""")

Key Features

9 Tools: configure, launch, invoke, agents, status, logs, memory, identity, session
Zero DevOps: Build & deploy via CodeBuild with ARM64 containers
Self-Deploying Agents: Agents can autonomously deploy themselves
Memory Management: Support for STM (Short-Term Memory) and LTM (Long-Term Memory)
OAuth Integration: Built-in identity provider management

Available Tools

Tool	Purpose
`configure`	Generate Dockerfile, IAM roles, config YAML
`launch`	Build & deploy to AgentCore via CodeBuild
`invoke`	Execute deployed agent with payload
`agents`	List/get agent runtimes
`status`	Check agent health & endpoint status
`logs`	Query CloudWatch logs
`memory`	Manage AgentCore memories (STM/LTM)
`identity`	OAuth provider management
`session`	Stop active runtime sessions

Configuration

Requires AWS credentials with permissions for:

bedrock-agentcore:*
ecr:*
codebuild:*
iam:*
logs:*

Resources