Skills
Navigate
skills/cekura-ai/cekura-skills/cekura-personality-design/Gen Agent Trust Hub

cekura-personality-design

Pass

Audited by Gen Agent Trust Hub on Oct 23, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill manages personality prompts and background noise URLs, which are external data ingestion points. \n
  • Ingestion points: The persona prompt and background_noise URL fields within the personalities_create and personalities_partial_update tools (documented in SKILL.md and BUNDLE.md). \n
  • Boundary markers: The skill establishes a clear conceptual boundary between Personality (audio layer) and Evaluator (task logic) to prevent instruction confusion. \n
  • Capability inventory: Platform tools such as personalities_create, personalities_partial_update, personalities_list, and personalities_retrieve are used to manage these settings. \n
  • Sanitization: Content validation is handled by the Cekura platform tools; the skill provides design guidance for proper configuration.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 23, 2026, 04:05 AM