short-video-ads-master
Pass
Audited by Gen Agent Trust Hub on Jul 1, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The core instructions (Master OS and Sub-skills) provide industry-specific mental models and playbooks for advertising strategy. No prompt injection, DAN-style overrides, or safety bypass attempts were identified.
- [SAFE]: The interactive Bash scripts located in the
cli/directory (e.g.,agentic.sh,workflow/50.sh) facilitate workflow walkthroughs and report generation. These scripts are designed for local execution by the user and operate entirely on standard input/output. They do not perform unauthorized network operations, sensitive file system access, or privilege escalation. - [SAFE]: The 'Agentic Protocol' instructs the agent to use its own research tools to gather information from industry-standard platforms (such as Ocean Engine or ChanMama). This is consistent with the skill's primary purpose and does not involve malicious redirections or unsafe data processing.
- [SAFE]: No obfuscated content, multi-layered encoding, hidden URLs, or hardcoded credentials were found across the skill's files. Dependencies mentioned in the documentation are well-known industry tools.
Audit Metadata