skills/lev-os/agents/prompt-engineer/Gen Agent Trust Hub

prompt-engineer

Pass

Audited by Gen Agent Trust Hub on Mar 1, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [NO_CODE]: The skill is entirely composed of Markdown text and YAML metadata. It contains no scripts (.py, .js, .sh), binaries, or executable commands that could interact with the host system.
  • [SAFE]: No malicious patterns such as prompt injection (to bypass safety), data exfiltration, or obfuscation were detected. The skill's 'magic mode' is a behavioral instruction for the agent to operate without meta-commentary, not a safety bypass mechanism. The instructions focus exclusively on legitimate prompt optimization techniques using established frameworks like RTF and STAR.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 1, 2026, 09:43 AM