AWS prescriptive-guidance documentation change
Summary
Added 'platforms' component to AI framework model and expanded protocol coverage
Security assessment
Added content about platforms handling security boundaries and safe interactions. Specifically mentions security features like identity management and secure tool interactions.
Diff
diff --git a/prescriptive-guidance/latest/agentic-ai-frameworks/conclusion.md b/prescriptive-guidance/latest/agentic-ai-frameworks/conclusion.md index d9a023a88..7604f0f51 100644 --- a//prescriptive-guidance/latest/agentic-ai-frameworks/conclusion.md +++ b//prescriptive-guidance/latest/agentic-ai-frameworks/conclusion.md @@ -3 +3 @@ -[Documentation](/index.html)[AWS Prescriptive Guidance](https://aws.amazon.com/prescriptive-guidance/)[Agentic AI frameworks, protocols, and tools on AWS](introduction.html) +[Documentation](/index.html)[AWS Prescriptive Guidance](https://aws.amazon.com/prescriptive-guidance/)[Agentic AI frameworks, platforms, protocols, and tools on AWS](introduction.html) @@ -7 +7 @@ -The landscape of agentic AI continues to evolve rapidly, offering organizations powerful new ways to build intelligent, autonomous systems. This guide has explored three essential components for successful implementation: _frameworks_ that provide the foundation, _protocols_ that enable communication, and _tools_ that extend capabilities. +The landscape of agentic AI continues to evolve rapidly, offering organizations powerful new ways to build intelligent, autonomous systems. This guide has explored three essential components for successful implementation: _frameworks_ that provide the foundation, _platforms_ that provide the environment, _protocols_ that enable communication, and _tools_ that extend capabilities. @@ -10,0 +11,2 @@ As frameworks mature, you can expect increased interoperability, standardization +Platforms provide the execution, governance, and lifecycle environment in which agentic systems operate. They handle concerns such as identity, security boundaries, observability, memory management, session grounding, and safe interaction with tools and data. In AWS environments, platforms like managed agent runtimes and orchestration services allow organizations to deploy, monitor, evolve, and govern autonomous agents and agentic systems at scale. Platforms bridge foundational frameworks with real-world operational requirements. + @@ -13 +15 @@ The choice of agent protocols represents a strategic decision that balances imme -For most organizations, MCP represents a strong foundation due to its open standard, growing ecosystem, support for agent-to-agent communication patterns, and tool integration capabilities. AWS has embraced MCP as a strategic protocol, actively contributing to its development and implementing it across services like the [Strands Agents SDK](https://strandsagents.com/latest/). By using MCP alongside appropriate framework-native tools and meta-tools, you can build agent systems that deliver immediate value while remaining adaptable to future innovations. +For most organizations, MCP represents a strong foundation due to its open standard, growing ecosystem, support for agent-to-agent communication patterns, and tool integration capabilities. AWS has embraced MCP and Agent2Agent (A2A) as strategic protocols, actively contributing to their development and implementing them across services like the [Strands Agents SDK](https://strandsagents.com/latest/). By using MCP or A2A alongside appropriate framework-native tools and meta-tools, you can build agent systems that deliver immediate value while remaining adaptable to future innovations.