subagent-driven-development
Pass
Audited by Gen Agent Trust Hub on Feb 27, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill defines a logical and secure workflow for software development orchestration. It promotes best practices such as Test-Driven Development (TDD) and separation of concerns through the use of fresh subagent contexts per task.
- [SAFE]: Analysis of the prompt templates confirms that all instructions are directed toward task performance and quality assurance. There are no attempts to bypass safety filters, extract system prompts, or disregard core agent constraints.
- [SAFE]: No hardcoded credentials, sensitive file paths, or suspicious network operations (e.g., use of curl or wget to untrusted domains) were found within the skill files or templates.
- [SAFE]: The skill integrates with a local ecosystem of 'superpowers' skills for specialized operations, which is a standard pattern for modular AI agent functionality and does not involve unverifiable external dependencies.
- [SAFE]: While the skill processes plan files that could technically serve as an ingestion point for indirect instructions, the multi-stage review process (spec compliance then code quality) acts as a functional guardrail against unintended behavior during execution.
Audit Metadata