copywriting
Pass
Audited by Gen Agent Trust Hub on Aug 3, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [SAFE]: The skill and its associated scripts do not exhibit malicious patterns such as obfuscation, credential theft, or unauthorized network access. The functionality aligns with the stated purpose of improving copywriting quality through structured methodologies and automated linguistic analysis.\n- [COMMAND_EXECUTION]: The skill includes a local Node.js utility (
resources/copy-scorer.js) for text analysis. This tool uses standard Node.js APIs to read and process text or files provided by the user. The code uses regular expressions for pattern matching and does not perform any remote code execution or shell interpolation.\n- [SAFE]: Assessment of indirect prompt injection surfaces:\n - Ingestion points: Text strings or file paths provided to the
copy-scorer.jsutility for linguistic scoring.\n - Boundary markers: The workflow instructions provide a structured context for the agent to generate and evaluate copy, though no explicit delimiter-based sanitization is implemented in the tool itself.\n
- Capability inventory: Local script execution and file reading via Node.js for analysis purposes.\n
- Sanitization: Input text is analyzed using static regular expressions and linguistic heuristics for word counts and syllables; it is not evaluated as code or interpreted as agent instructions.
Audit Metadata