copywriting-prose-creator

Pass

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides specific Python and Bash script templates in references/audit-tools.md for the agent to use when evaluating a content corpus. These scripts calculate statistical prose metrics such as mean sentence length and n-gram frequency.
  • [PROMPT_INJECTION]: The skill ingests untrusted external content (a 'corpus') to extract prose patterns, which creates a surface for indirect prompt injection where instructions hidden in the corpus could attempt to influence the agent's behavior. 1. Ingestion points: Content corpus from local folders or external URLs processed in AUDIT mode. 2. Boundary markers: No explicit delimiters or instructions are provided to the agent to ignore embedded commands within the corpus. 3. Capability inventory: Significant capabilities including Read, Edit, Write, WebFetch, and the Agent tool for spawning sub-agents. 4. Sanitization: No evidence of sanitization or filtering of corpus content prior to analysis.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 13, 2026, 12:50 AM
Security Audit — agent-trust-hub — copywriting-prose-creator