Back to skill
Skillv1.0.0
ClawScan security
Frontend Design Anthropic · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
BenignMay 1, 2026, 7:11 AM
- Verdict
- Benign
- Confidence
- high
- Model
- gpt-5.5
- Summary
- This is an instruction-only frontend design skill with no code, install steps, credentials, or broad system access requested.
- Guidance
- This skill appears safe to install as a design-guidance skill. As with any code-generation aid, review generated frontend code for accessibility, performance, licensing of fonts/assets, and production suitability before deployment.
Review Dimensions
- Purpose & Capability
- okThe instructions are coherent with the stated purpose of helping create distinctive frontend interfaces and production-grade UI code.
- Instruction Scope
- okThe skill gives aesthetic and implementation guidance for frontend generation; it does not direct the agent to override user goals, hide actions, or perform unrelated tasks.
- Install Mechanism
- okThere is no install spec and no code files; the artifact set is instruction-only.
- Credentials
- okNo required binaries, environment variables, credentials, network access, or local file access are declared.
- Persistence & Privilege
- okThe artifacts do not request persistence, background execution, elevated privileges, credential use, or persistent memory.