trimming-project-instructions
Pass
Audited by Gen Agent Trust Hub on Oct 30, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill contains purely instructional markdown designed to guide the agent in managing its own project instruction files (like CLAUDE.md).
- [SAFE]: No prompt injection, data exfiltration, or obfuscation patterns were detected.
- [SAFE]: The skill does not reference any external packages, remote scripts, or unverifiable dependencies.
- [SAFE]: Although the skill mentions mechanical extraction using scripts, it does not provide or load external executable code, nor does it encourage unsafe execution practices.
Audit Metadata