cognitive-scaffolding
Pass
Audited by Gen Agent Trust Hub on Apr 12, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill's primary function is prompt structure and context organization through string manipulation. No dangerous system calls, subprocess spawning, or unauthorized file system operations are present.
- [SAFE]: Code snippets provided in TypeScript and Python are helper utilities for managing context zones (Anchor, Foreground, Middle) and do not include remote code execution or data exfiltration logic.
- [SAFE]: External links to github.com/itallstartedwithaidea and googleadsagent.ai are vendor-owned resources for documentation and project reference, which do not represent a security risk.
- [SAFE]: No prompt injection patterns or attempts to override model safety constraints were detected in the instructions or provided code.
- [SAFE]: The skill implements structural best practices for context window management, such as the use of explicit XML-style delimiters, which serve as a mitigation against potential confusion between data and instructions.
Audit Metadata