Back to skill
Skillv1.0.0
ClawScan security
chinese-punctuation-correction · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
BenignMay 1, 2026, 8:04 AM
- Verdict
- Benign
- Confidence
- high
- Model
- gpt-5.5
- Summary
- This is a coherent instruction-only Chinese punctuation correction skill with no code, credentials, install steps, persistence, or broad system access shown.
- Guidance
- This skill appears safe to install for proofreading Chinese punctuation. Users should still review corrected text for meaning, since the skill is designed to edit user-provided writing.
Review Dimensions
- Purpose & Capability
- okThe stated purpose is punctuation-focused Chinese sentence correction, and the included instructions and reference rules are aligned with that purpose.
- Instruction Scope
- okThe skill asks the agent to preserve meaning, make minimal edits, and only provide explanations when requested, which keeps the behavior bounded.
- Install Mechanism
- okNo install specification, code files, binaries, packages, or environment variables are declared.
- Credentials
- okThe artifacts do not request local file access beyond the included reference document, network access, credentials, shell commands, or external tools.
- Persistence & Privilege
- okNo persistent storage, background behavior, elevated privileges, account access, or cross-session memory use is described.
