build-skill
Pass
Audited by Gen Agent Trust Hub on Apr 13, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill uses the Agent and Skill tools for legitimate orchestration of development tasks.
- [COMMAND_EXECUTION]: Employs the Agent tool to run parallel background tasks and creates temporary working directories in
/tmp/skill-compare/. These operations are well-defined and aligned with the skill's stated purpose of building and comparing skill variations. - [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied specifications to generate new skill content. While this is an ingestion surface for external data, the synthesis and comparison steps provide a layer of agent review before the final skill is written to the persistent skill directory (
~/.claude/skills/). - [EXTERNAL_DOWNLOADS]: The skill references other internal or platform-specific skills (
skill-creatorandsuperpowers) to assist in the generation process. It does not perform unauthorized network downloads or execute external scripts from untrusted sources.
Audit Metadata