flutter-sentry-engineering

Pass

Audited by Gen Agent Trust Hub on Oct 24, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill suggests using the rg (ripgrep) utility to search within its own references/production-guide.md file for specific documentation sections. This is a standard and safe use of search tooling for local content management.
  • [EXTERNAL_DOWNLOADS]: The guide includes instructions for installing official Sentry packages from the pub.dev registry (specifically sentry_flutter, sentry_dio, and sentry_dart_plugin). These are standard, well-known libraries within the Flutter ecosystem.
  • [INDIRECT_PROMPT_INJECTION]: The skill facilitates the capture and processing of application events and logs which may contain untrusted user data. The documentation proactively addresses this by providing mandatory guidance on privacy controls and PII masking. 1. Ingestion points: captureException, SentryLog, and Session Replay as described in references/production-guide.md. 2. Boundary markers: Includes instructions to implement beforeSend and beforeBreadcrumb hooks to filter telemetry. 3. Capability inventory: The skill involves network transmission of telemetry data to Sentry's ingestion services. 4. Sanitization: Provides comprehensive rules for PII redaction, credential avoidance, and visual masking using SentryMask and options.privacy.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 24, 2026, 11:54 PM