content-strategist
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- Indirec t Promp t Injection (LOW): The skill documentatio n identifie s work flow s tha t inges t untruste d data, potentially allowin g indirec t injectio n at tack s throug h proces s e d file s. * Ingestio n point s: script s/voice_analyzer.p y proces s e s article.m d and script s/content_audit.p y proces s e s sitemap.xml. * Boundary marker s: Absen t in documentation. * Capabilit y inventor y: Loca l Pytho n scrip t executio n fo r analysi s and reporting. * Sanitization: No mentio n of inpu t validatio n or filterin g fo r externa l file s.
- ComMand Executio n (SAFE): The skill include s instruction s fo r runnin g interna l utilit y script s. * Evidence: Bas h block s in SKIL L.m d reference script s like script s/content_report.p y whic h are intende d fo r legitimate functiona l use.
Audit Metadata