think-forge
Pass
Audited by Gen Agent Trust Hub on Sep 5, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [NO_CODE]: The skill consists entirely of markdown instructions and YAML configuration. It contains no executable scripts, shell commands, or external dependencies.
- [SAFE]: The skill's instructions strictly enforce a 'read-only boundary' and 'stop conditions.' It is designed to analyze requests and map them to appropriate domain owners (e.g.,
auth-access,data-storage) without actually performing the work or inspecting the file system. - [PROMPT_INJECTION]: While the skill contains instructional language about how the AI should behave (e.g., 'THESE RULES MUST BE FOLLOWED'), these are legitimate operational constraints for the skill's purpose and do not attempt to bypass platform safety filters or exfiltrate data.
- [DYNAMIC_CONTEXT_INJECTION]: There are no instances of dynamic context injection (the
!command`` pattern) in the SKILL.md or reference files.
Audit Metadata