tighten-instruction
Pass
Audited by Gen Agent Trust Hub on Aug 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill consists entirely of natural language instructions for stylistic text refinement. It does not include executable code, external dependencies, or network capabilities.
- [INDIRECT_PROMPT_INJECTION]: The skill processes instructions from external files (e.g., CLAUDE.md). While this is an ingestion point for untrusted data, the skill focuses on linguistic clarity and includes a self-scoring step to ensure original meaning is preserved.
Audit Metadata