impeccable
Fail
Audited by Gen Agent Trust Hub on Nov 10, 2026
Risk Level: HIGHEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill's launcher scripts (scripts/impeccable and scripts/impeccable.cmd) are configured to fetch a platform-specific binary engine from an external GitHub repository (github.com/pbakaus/impeccable) when a local version is unavailable.
- [REMOTE_CODE_EXECUTION]: The skill executes the downloaded binary engine as its primary analysis tool. Executing unverified binaries from third-party sources not associated with the author or trusted vendors represents a supply-chain risk.
- [COMMAND_EXECUTION]: The skill uses extensive shell commands, including curl and wget for network downloads, chmod for permission changes, and mobile development tools like adb (Android) and xcrun simctl (iOS) for device interaction.
- [PROMPT_INJECTION]: The instructions for degraded sub-agents and specialized agent roles use command-like phrases such as 'Step fully out of the work you just finished' and 'adopt only this file's instructions' to enforce behavioral overrides and context isolation.
Recommendations
- AI detected serious security threats
Audit Metadata