Back to skill
Skillv1.0.0
ClawScan security
笔杆子写稿 · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
BenignMay 1, 2026, 8:25 AM
- Verdict
- Benign
- Confidence
- high
- Model
- gpt-5.5
- Summary
- This is a simple instruction-only writing skill that asks the agent to draft polished Chinese formal text and does not request tools, credentials, installation, persistence, or data access.
- Guidance
- This skill appears low risk from the provided artifacts. Before using it, provide only the information needed for the draft and review the generated text for accuracy, originality, and appropriateness before publishing or submitting it.
Review Dimensions
- Purpose & Capability
- okThe stated purpose and SKILL.md content are coherent: generate complete, fluent, original formal drafts based on user needs.
- Instruction Scope
- okInstructions are limited to writing style and output format; they do not redirect user goals, demand unsafe tool use, or override external controls.
- Install Mechanism
- okNo install spec, binaries, dependencies, environment variables, or code files are present.
- Credentials
- okThe skill does not request local file access, network access, credentials, account permissions, or other environmental authority.
- Persistence & Privilege
- okNo persistence, background behavior, memory use, credential handling, or privilege escalation is evidenced in the artifacts.