app-marketing-copy
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION] (SAFE): The skill utilizes a local script
scripts/check_app_store_limits.pyto validate character counts of generated copy. The script is self-contained, uses standard Python libraries (argparse, json, sys, typing), and does not perform any network operations or unsafe system calls. - [DATA_EXFILTRATION] (SAFE): No hardcoded credentials, sensitive file path access, or network-capable commands (like curl or wget) were found. Data processing remains local to the user's environment.
- [PROMPT_INJECTION] (SAFE): The instructions in
SKILL.mdare focused on the intended marketing use case and do not contain patterns aimed at bypassing AI safety constraints, role-playing as unrestricted entities, or leaking system prompts. - [REMOTE_CODE_EXECUTION] (SAFE): The skill does not download or execute code from external sources. No package managers (pip, npm) or remote scripts are invoked.
- [INDIRECT_PROMPT_INJECTION] (SAFE): The skill has a low attack surface for indirect injection. 1. Ingestion points: User-provided app descriptions and JSON metadata files for limit checking. 2. Boundary markers: The skill uses clear intake templates to separate user inputs. 3. Capability inventory: Limited to text generation and a local character-counting script; no network or file-write capabilities. 4. Sanitization: The local script uses a strict JSON parser for input.
Audit Metadata