drama-evaluator

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [NO_CODE] (SAFE): The skill consists entirely of Markdown files and reference materials. There are no Python scripts, Node.js code, or shell commands included.
  • [EXTERNAL_DOWNLOADS] (SAFE): No external resources or packages are referenced or downloaded during the operation of this skill.
  • [DATA_EXFILTRATION] (SAFE): The skill does not possess any network access tools or data transmission capabilities.
  • [PROMPT_INJECTION] (SAFE): Instructions are strictly functional and focused on the evaluation task, with no attempts to bypass LLM safety guardrails.
  • [INDIRECT_PROMPT_INJECTION] (SAFE): Although the skill processes user-provided drama scripts, it has no tool access (allowed-tools is empty), which eliminates any potential for adversarial data to cause external harm.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:32 PM