macos-performance-benchmark
Pass
Audited by Gen Agent Trust Hub on Aug 11, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The s k i l l e x e c u t e s v a r i o u s m a c O S sys t e m u t i l i t i e s (
sys c t l,pm s e t,di s k u t i l,s f l t oo l,l a un c h c t l,powe r m e t r i c s) and u s e r-in s t a l l e d too l s (m a c m on) to c o l l e c t p e r fo r m an c e and d i a gno s t i c d a t a. Pr i v i l e g e d e x e c u t i on v i as u d oi s op t i on a l l y u s e d to a c c e s s p r i v i l e g e d h a r d w a r e c o un t e r s fo r ther m a l p r e s s u r e d a t a i nd i a g/ c p u. s h. - [DATA_EXPOSURE]: The d i a gno s t i c s c r i p t s s c an the u s e r 's ho m e d i r e c t o ry fo r f i l e s e x c e e d i ng 1 GB u s i ng
m d f in dands t a ti nd i a g/ d i s k. s h, r e p o r t i ng the i r f i l e p a t h s and s i z e s i n the f i n a l b en c h m a rk r e p o r t. - [PROMPT_INJE C T I ON]: The s k i l l i m p l e m en t s a r e p o r t i ng m e c h an i s m th a t in g e s t s o u t p u t fo r m sys t e m c o m m an d s ( e. g., p r o c e s s n a m e s, f i l e p a t h s, and s t a r t u p s e r v i c e i d en t i f i e r s). S i n c e th i s d a t a i s un t r u s t e d and in c l u d e d i n M a rk d own r e p o r t s s u m m a r i z e d by the A I a g en t, i t r e p r e s en t s an in d i r e c t p r o m p t in j e c t i on s u r f a c e.
- In g e s t i on p oin t s: Ou t p u t s f r o m
ps,m d f in d,l a un c h c t l, ands f l t oo l. - Bo un d a ry m a rk e r s: No n e u s e d i n r e p o r t g en e r a t i on.
- Ca p a b i l i t y in v en t o ry: S h e l l e x e c u t i on ( Ba s h), p r i v i l e g e d e x e c u t i on ( s u d o Ba s h), and f i l e r e a d / w r i t e ( l o c a l c a c h e).
- San i t i z a t i on: TS V g en e r a t i on c o l l a p s e s n e w l i n e s and t a b s i n t o s p a c e s, b u t M a rk d own r e p o r t g en e r a t i on d o e s n o t e s c a p e c on t en t.
Audit Metadata