skills/webpro/skills/review-prose/Gen Agent Trust Hub

review-prose

Pass

Audited by Gen Agent Trust Hub on Nov 12, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill directs the agent to run a bundled Node.js script (scripts/llm-cliche-highlighter/llm-cliche-highlighter.mjs) for prose analysis. This script is a local utility that uses regular expressions to identify stylistic patterns; it does not perform network operations or access sensitive system files.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external prose as part of its core functionality (e.g., documentation and README files). This presents a standard surface for indirect prompt injection. The skill includes explicit instructions for the agent to maintain source fidelity and follow a verification checklist, which serves as a defensive boundary for the information being processed.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 12, 2026, 08:02 PM