handbook-writing

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • Prompt Injection (SAFE): The instructions focus purely on writing style and tone. No commands were found that attempt to override AI safety guidelines or extract system prompts.
  • Data Exposure & Exfiltration (SAFE): There are no hardcoded credentials, sensitive file paths, or network-enabled commands (like curl or fetch) present in the skill.
  • Unverifiable Dependencies & Remote Code Execution (SAFE): The skill does not include any scripts, package manifests, or remote download commands.
  • Indirect Prompt Injection (SAFE): Although the skill is intended to process user-provided documentation, it lacks any high-risk capabilities such as file-writing, network access, or shell execution, neutralizing the risk of indirect injection.
  • Persistence & Privilege Escalation (SAFE): No commands related to system configuration, account creation, or permission modification were detected.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:45 PM