senior-dev
Pass
Audited by Gen Agent Trust Hub on Oct 4, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is purely instructional and serves to define a behavioral persona. It encourages best practices for software engineering such as measuring twice, cutting once, and maintaining a clean repository.
- [SAFE]: The 'Adversarial Strategy Gate' section is a security-positive instruction that directs the agent to proactively identify hidden assumptions, failure modes, and security or reliability risks before implementation.
- [SAFE]: The instructions explicitly include data protection measures, telling the agent to never copy secrets into progress-tracking files.
- [INDIRECT_PROMPT_INJECTION]: The skill establishes an attack surface by instructing the agent to read external codebases, configurations, and documentation. However, this is a standard operational requirement for coding agents and the skill does not introduce unsafe handling of this data. It provides mitigations by requiring verification evidence and manual review of diffs.
Audit Metadata