error-tracking
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- Prompt Injection (SAFE): No instructions found that attempt to bypass safety filters or override system prompts.
- Data Exposure & Exfiltration (SAFE): The skill correctly uses environment variables for the Sentry DSN and explicitly warns against logging sensitive data in error contexts. Network operations are limited to local testing and standard Sentry reporting.
- Unverifiable Dependencies (SAFE): The skill references standard, well-known Sentry packages (@sentry/node).
- Indirect Prompt Injection (LOW): The skill handles untrusted data via error objects and requests, but includes explicit instructions to avoid leaking sensitive data, mitigating risk.
- Command Execution (SAFE): curl commands are provided only for testing local development endpoints.
Audit Metadata