Data Privacy
Learn how to control data collection and protect sensitive information in AI agent monitoring.
By default, Sentry AI agent integrations respect your SDK's PII settings:
sendDefaultPii: false- Only metadata like model names, token counts, tool names, and execution times are collectedsendDefaultPii: true- Full inputs, outputs, prompts, and responses from AI models and tools are captured.
You can override this by configuring it on a function call level. For more details see Configuration.
Sentry has built-in Server-Side Data Scrubbing that automatically detects and removes common sensitive data like credit card numbers, social security numbers, and passwords from most event data.
However, the following AI agent span attributes are not protected by default:
gen_ai.promptgen_ai.request.messagesgen_ai.tool.inputgen_ai.tool.outputgen_ai.response.tool_callsgen_ai.response.textgen_ai.response.object
If you wish to enable Data Scrubbing for any of these fields you can add them to Global Sensitive Fields in Security & Privacy settings or to Additional Sensitive Fields in Project settings under Security & Privacy.
Was this helpful?
Help improve this content
Our documentation is open source and available on GitHub. Your contributions are welcome, whether fixing a typo (drat!) or suggesting an update ("yeah, this would be better").
Our documentation is open source and available on GitHub. Your contributions are welcome, whether fixing a typo (drat!) or suggesting an update ("yeah, this would be better").