skills/im5tu/claude/copy-editing/Gen Agent Trust Hub

copy-editing

Pass

Audited by Gen Agent Trust Hub on Feb 26, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is designed to process and edit user-provided marketing copy, which functions as an ingestion point for untrusted data.
  • Ingestion points: Marketing copy provided by the user is processed through multiple editing sweeps as defined in SKILL.md.
  • Boundary markers: The framework does not define specific delimiters or instructions to prevent the agent from obeying instructions embedded within the user's copy.
  • Capability inventory: Analysis of SKILL.md and references/plain-english-alternatives.md shows no evidence of dangerous capabilities such as subprocess calls, file-write operations, or network requests.
  • Sanitization: There are no specific mechanisms described to sanitize or filter potential injection strings from the input copy.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 26, 2026, 11:37 PM