dio-refine-prompt

Pass

Audited by Gen Agent Trust Hub on Jul 23, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is designed for interactive chat-based prompt refinement. It contains explicit instructions to avoid tool usage, file modifications, or network activities, stating: 'no tools, no file edits, no executing downstream work' and 'Do not run tools, read files, or fetch URLs'.
  • [SAFE]: The YAML frontmatter includes disable-model-invocation: true, which restricts the execution environment from making unauthorized external model calls.
  • [SAFE]: No patterns of prompt injection, obfuscation, or data exfiltration were found. The skill operates on user-provided text drafts and produces structured text output without interacting with the host system or external services.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 23, 2026, 01:01 AM
Security Audit — agent-trust-hub — dio-refine-prompt