Product Description Writer
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFENO_CODE
Full Analysis
- PROMPT_INJECTION (SAFE): No instructions to bypass safety filters or override system behavior were detected. All triggers and workflows are focused on product copywriting.- DATA_EXFILTRATION (SAFE): The skill does not perform any network operations, API calls, or file system access.- REMOTE_CODE_EXECUTION (SAFE): There are no external dependencies, remote script downloads, or execution patterns found in the file.- COMMAND_EXECUTION (SAFE): No shell commands, subprocesses, or system-level scripts are present.- NO_CODE (SAFE): The skill consists entirely of markdown documentation and structural triggers for text generation. No code is shipped or executed.- INDIRECT_PROMPT_INJECTION (SAFE): While the skill processes user-provided product data, it lacks any dangerous capabilities (file writes, network, execution) to exploit. Ingestion points: Product details in triggers. Boundary markers: Absent. Capability inventory: None. Sanitization: Absent.
Audit Metadata