Skyflow interceptor protects agent tool calls
Understand this story
The AWS Machine Learning Blog introduced a solution: Skyflow integrates with Amazon Bedrock AgentCore Gateway to de-identify and restore sensitive data before and after an agent calls a tool, via request and response interceptors, so that sensitive data neither enters the tools the agent calls nor enters the model context those tools feed back. Skyflow is available on AWS Marketplace, and the interceptor is open source on GitHub. String de-identification and restoration are generally available; document de-identification via signed URLs uses the preview of Skyflow Detect v2. AgentCore Gateway availability varies by AWS Region.
AI overview based on source reports · Updated 17 hr ago
Coverage timeline
Follow the coverage to explore different perspectives.
- AWS Machine Learning BlogProtect agent tool calls with Skyflow and Amazon Bedrock AgentCore Gateway
The post describes adding Skyflow Detect as a request and response interceptor on Amazon Bedrock AgentCore Gateway so sensitive data is de-identified before reaching agent tools and restored before the agent reads results, with documents replaced by de-identified copies. It details three modes—Shield the tool, Shield the model, and Shield the document—and notes the interceptor is open source on GitHub under the MIT license, with Skyflow available in AWS Marketplace. It also covers deployment via AWS CDK, fail-closed behavior, and considerations such as PDFs and images receiving one-way tokens and Skyflow Detect v2 being in preview.
Activity for this story
- Current comparable activity
- 6
- Peak comparable activity
- 10Nov 20 03:00
- Change in the past 24 hours
- –
Trends compare the same fully observed participants, which may cover fewer participants than current activity. Move the pointer or tap the chart to inspect hourly activity; use the left and right arrow keys with a keyboard.