DeAI
Pass
Audited by Gen Agent Trust Hub on Jun 14, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is composed strictly of Markdown documentation and prompt instructions. There are no executable scripts (.py, .js, .sh), binaries, or configuration files that could execute code on a host system.
- [SAFE]: No data exfiltration or unauthorized network operations were detected. The skill specifically states it does not make network requests, and analysis confirms the absence of tools or commands (like curl or wget) that would enable network activity.
- [SAFE]: All external links provided in the metadata and body (ilang.ai and github.com/ilang-ai/*) are consistent with the identified author 'ilang-ai' and represent legitimate vendor resources.
- [SAFE]: The skill's instructions focus on text transformation (removing filler words, adjusting rhythm). It includes explicit constraints to prevent the agent from generating fabricated content or slang, requiring the human user to provide authentic input via placeholders.
Audit Metadata