sentry-flutter-sdk

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill recommends installing the official sentry-wizard utility via Homebrew from the vendor's repository (getsentry/tools/sentry-wizard). This is a standard and safe practice for the primary purpose of the skill.
  • [COMMAND_EXECUTION]: During the detection phase, the skill executes common shell commands such as cat, grep, ls, and find to identify the Flutter project type, existing dependencies (like dio, go_router, sqflite), and target platforms. These commands are non-privileged and used for context discovery.
  • [INDIRECT_PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection as it ingests and processes content from local project files (e.g., pubspec.yaml, source code) to guide its setup recommendations.
  • Ingestion points: Reads pubspec.yaml, lib/main.dart, and directory structures using grep and ls in SKILL.md Phase 1.
  • Boundary markers: None explicitly used for the output of discovery commands.
  • Capability inventory: Uses discovery data to recommend flutter pub add commands and configuration changes to main.dart and pubspec.yaml (seen in SKILL.md Path B).
  • Sanitization: No specific sanitization of the file content is performed before presenting setup options to the user.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 07:09 AM
Security Audit — agent-trust-hub — sentry-flutter-sdk