Security audit
Security checks for vulnerabilities and agentic risk
This skill is a plain guidance package for improving skill metadata and discoverability, with no executable code or hidden runtime behavior found.
This appears safe to install as a metadata and listing optimization aid. Users should treat its advice as discoverability guidance, not as a general skill authoring, security review, or publication approval authority.
Skill contains instructions that could directly expose system prompts, internal rules, or hidden instructions to users or external parties.
- 3 to 8 realistic example prompts
- one short note on what not to claim in the listing
## Output Rules
- Keep names literal, ASCII-safe, and easy to type.
- Put the highest-signal trigger terms early in the description.
Skill modifies its own code, configuration, or behavior at runtime. Self-modification enables an agent to escalate privileges, disable safety constraints, or install persistent backdoors.
- `Check whether this TikTok Shop video is compliant`
- `Compare this exported ad against the product listing image`
- `Rewrite this skill description so agents can find it more easily`
## Scope discipline
The skill description is extremely broad and can trigger on generic tasks like creating, refining, packaging, or publishing a skill, rather than a narrowly bounded SEO/discoverability use case. In agent ecosystems, overly broad invocation language can cause misrouting, unnecessary activation, and inappropriate trust in this skill for tasks outside its competence, which can degrade decision quality or override more suitable specialized skills.
No suspicious patterns detected.