According to AWS Machine Learning, policies are expressed in Dogwood, which AWS Machine Learning describes as an open source governance language.
According to AWS Machine Learning, the Dogwood monitor applies these policies to agent actions in real time.
According to AWS Machine Learning, the Dogwood monitor is built into AgentCore Gateway.
According to AWS Machine Learning, AgentCore Gateway is a capability of Amazon Bedrock AgentCore.
According to AWS Machine Learning, this briefing reflects AWS Machine Learning's description only; the provided evidence does not establish independent confirmation or additional details about the policies, monitoring behavior, or product operation.
Sources
- [Authoring Dogwood policies from natural language in Amazon Bedrock AgentCore](https://aws.amazon.com/blogs/machine-learning/authoring-dogwood-policies-from-natural-language-in-amazon-bedrock-agentcore/)