absolute-human
Pass
Audited by Gen Agent Trust Hub on Apr 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes shell commands for environment discovery and quality assurance. It performs directory listings to detect available companion skills and invokes standard project commands such as
npm test,make, orpytestto verify that implementation tasks satisfy the defined acceptance criteria. - [EXTERNAL_DOWNLOADS]: The skill instructions and README recommend downloading additional specialized engineering skills from the AbsolutelySkilled GitHub organization using the
npx skills addcommand. - [PROMPT_INJECTION]: The skill defines an indirect prompt injection surface as it processes external project data and user requirements during its intake and discovery phases. This data is used to generate execution plans and drive agent actions. The framework manages this risk through strict phase boundaries and mandatory user approval checkpoints before any code changes are executed.
Audit Metadata