app-marketing-copy
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: Technical analysis of the skill instructions and associated files found no evidence of malicious behavior, obfuscation, or unauthorized data access.
- [COMMAND_EXECUTION]: The skill references a local helper script
scripts/check_app_store_limits.pyto validate the character counts of generated copy. A manual review of the Python script confirms it is a benign utility that uses standard libraries (json,argparse,sys) to perform string processing. It does not contain network operations or functionality to access sensitive files. - [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied application names and features to generate marketing content, creating a standard data ingestion surface.
- Ingestion points: SKILL.md (User input intake template and examples)
- Boundary markers: Absent; the skill does not use specific delimiters to separate user data from instructions.
- Capability inventory: The skill provides a local script for metadata length validation; it does not request network access or provide tools for file system modification.
- Sanitization: None; input is interpolated directly into templates. The risk is assessed as safe given the lack of dangerous capabilities associated with the skill's specific purpose.
Audit Metadata