prompt-enhancer
Pass
Audited by Gen Agent Trust Hub on Mar 1, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill inherently processes untrusted user input to generate instructions that are then executed by the agent. This functional flow represents a surface for indirect prompt injection.
- Ingestion points: The user input following the 'p:' or 'prompt:' trigger in SKILL.md is processed as the raw intent.
- Boundary markers: The skill does not currently use specific delimiters or explicit safety warnings to wrap the raw user intent during processing.
- Capability inventory: The skill is designed to rewrite and then directly execute the enhanced prompt.
- Sanitization: No sanitization or filtering of the user-provided text is performed prior to prompt generation.
Audit Metadata