writing-analyzer

Pass

Audited by Gen Agent Trust Hub on Mar 6, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill operates entirely within the LLM's internal processing context. It does not contain scripts, external dependencies, or instructions to execute shell commands or access the local file system.
  • [PROMPT_INJECTION]: The skill was evaluated for indirect prompt injection risks. Although it processes user-provided text for analysis, it does not possess any dangerous capabilities that could be exploited via malicious input.
  • Ingestion points: User-supplied Chinese and English text for stylistic analysis.
  • Boundary markers: None present in the prompt instructions.
  • Capability inventory: No file operations, network requests, or subprocess execution.
  • Sanitization: None; input is analyzed directly by the LLM.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 6, 2026, 04:19 PM