AWS Security ChangesHomeSearch

AWS bedrock-agentcore documentation change

Service: bedrock-agentcore · 2026-04-01 · Documentation low

File: bedrock-agentcore/latest/devguide/using-any-agent-framework.md

Summary

Removed Microsoft AutoGen and CrewAI framework documentation and code examples from integration guide

Security assessment

This change removes support documentation for two third-party frameworks but shows no evidence of security vulnerabilities or security-related reasons for removal. It appears to be a routine content update or deprecation of certain integration examples.

Diff

diff --git a/bedrock-agentcore/latest/devguide/using-any-agent-framework.md b/bedrock-agentcore/latest/devguide/using-any-agent-framework.md
index b806c93c8..84cfbdbfe 100644
--- a//bedrock-agentcore/latest/devguide/using-any-agent-framework.md
+++ b//bedrock-agentcore/latest/devguide/using-any-agent-framework.md
@@ -5 +5 @@
-Strands AgentsLangGraphGoogle ADKOpenAI Agents SDKMicrosoft AutoGenCrewAI
+Strands AgentsLangGraphGoogle ADKOpenAI Agents SDK
@@ -21,4 +20,0 @@ You can use open source AI frameworks to create an agent or tool. This topic sho
-  * Microsoft AutoGen
-
-  * CrewAI
-
@@ -221,144 +216,0 @@ For the full example, see [https://github.com/awslabs/amazon-bedrock-agentcore-s
-## Microsoft AutoGen
-
-For the full example, see [https://github.com/awslabs/amazon-bedrock-agentcore-samples/tree/main/03-integrations/agentic-frameworks/autogen](https://github.com/awslabs/amazon-bedrock-agentcore-samples/tree/main/03-integrations/agentic-frameworks/autogen).
-    
-    
-    from autogen_agentchat.agents import AssistantAgent
-    from autogen_agentchat.ui import Console
-    from autogen_ext.models.openai import OpenAIChatCompletionClient
-    import asyncio
-    import logging
-    
-    # Set up logging
-    logging.basicConfig(
-        level=logging.DEBUG,
-        format='%(asctime)s - %(name)s - %(levelname)s - %(message)s'
-    )
-    logger = logging.getLogger("autogen_agent")
-    
-    # Initialize the model client
-    model_client = OpenAIChatCompletionClient(
-        model="gpt-4o",
-    )
-    
-    # Define a simple function tool that the agent can use
-    async def get_weather(city: str) -> str:
-        """Get the weather for a given city."""
-        return f"The weather in {city} is 73 degrees and Sunny."
-    
-    # Define an AssistantAgent with the model and tool
-    agent = AssistantAgent(
-        name="weather_agent",
-        model_client=model_client,
-        tools=[get_weather],
-        system_message="You are a helpful assistant.",
-        reflect_on_tool_use=True,
-        model_client_stream=True,  # Enable streaming tokens
-    )
-    
-    # Integrate with Bedrock AgentCore
-    from bedrock_agentcore.runtime import BedrockAgentCoreApp
-    app = BedrockAgentCoreApp()
-    
-    @app.entrypoint
-    async def main(payload):
-        # Process the user prompt
-        prompt = payload.get("prompt", "Hello! What can you help me with?")
-        
-        # Run the agent
-        result = await Console(agent.run_stream(task=prompt))
-        
-        # Extract the response content for JSON serialization
-        if result and hasattr(result, 'messages') and result.messages:
-            last_message = result.messages[-1]
-            if hasattr(last_message, 'content'):
-                return {"result": last_message.content}
-        
-        return {"result": "No response generated"}
-    
-    app.run()
-    
-
-## CrewAI
-
-For the full example, see [https://github.com/awslabs/amazon-bedrock-agentcore-samples/blob/main/01-tutorials/01-AgentCore-runtime/01-hosting-agent/04-crewai-with-bedrock-model/runtime-with-crewai-and-bedrock-models.ipynb](https://github.com/awslabs/amazon-bedrock-agentcore-samples/blob/main/01-tutorials/01-AgentCore-runtime/01-hosting-agent/04-crewai-with-bedrock-model/runtime-with-crewai-and-bedrock-models.ipynb).
-    
-    
-    from crewai import Agent, Crew, Process, Task
-    from crewai_tools import MathTool, WeatherTool
-    from bedrock_agentcore.runtime import BedrockAgentCoreApp
-    import argparse
-    import json
-    app = BedrockAgentCoreApp()
-    
-    # Define CrewAI agent
-    def create_researcher():
-        """Create a researcher agent"""
-        from langchain_aws import ChatBedrock
-        
-        # Initialize LLM
-        llm = ChatBedrock(
-            model_id="anthropic.claude-3-sonnet-20240229-v1:0",
-            model_kwargs={"temperature": 0.1}
-        )
-        
-        # Create researcher agent
-        return Agent(
-            role="Senior Research Specialist",
-            goal="Find comprehensive and accurate information about the topic",
-            backstory="You are an experienced research specialist with a talent for finding relevant information.",
-            verbose=True,
-            llm=llm,
-            tools=[MathTool(), WeatherTool()]
-        )
-        
-    # Define the analyst agent
-    def create_analyst():
-    ....
-    
-    # Create the crew
-    def create_crew():
-        """Create and configure the CrewAI crew"""
-        # Create agents
-        researcher = create_researcher()
-        analyst = create_analyst()
-        
-        # Create research task with fields like description filled in as per crewAI docs
-        research_task = Task(
-            description="...",
-            agent=researcher,
-            expected_output="..."
-        )
-        
-        analysis_task = Task(
-        ...
-        )
-        
-        # Create crew
-        return Crew(
-            agents=[researcher, analyst],
-            tasks=[research_task, analysis_task],
-            process=Process.sequential,
-            verbose=True
-        )
-    
-    # Initialize the crew
-    crew = create_crew()
-    
-    # Finally write your entrypoint
-    @app.entrypoint
-    def crewai_bedrock(payload):
-        """
-        Invoke the crew with a payload
-        """
-        user_input = payload.get("prompt")
-        
-        # Run the crew
-        result = crew.kickoff(inputs={"topic": user_input})
-        
-        # Return the result
-        return result.raw
-    
-    if __name__ == "__main__":
-        app.run()
-