propose
Pass
Audited by Gen Agent Trust Hub on Oct 17, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [NO_CODE]: The skill consists entirely of markdown instructions and YAML templates. It defines logic for an agent to follow but contains no scripts (Python, JS, Bash) or binary executables.
- [SAFE]: All operations described (reading/writing
.lev/pm/tasks/, gathering context from conversation) are local to the project workspace and consistent with the stated purpose of task planning and alignment. - [SAFE]: The 'forward' mechanism in SKILL.md is a logical instruction to use another skill (
lev-plan) rather than a dangerous redirection or remote execution pattern.
Audit Metadata