Back to skill
Skillv0.1.0

ClawScan security

Transition Design · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

Scanner verdict

BenignMar 3, 2026, 11:00 AM
Verdict
benign
Confidence
high
Model
gpt-5-mini
Summary
The skill's requested resources and instructions match its stated purpose (designing frame-level transitions) and it does not request unrelated credentials, installs, or external endpoints.
Guidance
This skill appears internally consistent and low-risk: it only contains guidance and two reference KB files for calculating handles and selecting transitions. Before using it, ensure you supply the required clip metadata and frame-level assets (or another skill that provides them). Don’t grant unrelated credentials or broad filesystem access to the agent on behalf of this skill. If you plan to automate frame analysis, verify which tool will provide motion/lighting data and that those tools are trusted.

Review Dimensions

Purpose & Capability
okName/description (设计转场) align with the content: the skill only needs clip metadata, frame-level visual analysis, and the included KBs. No unrelated binaries, env vars, or config paths are required.
Instruction Scope
noteSKILL.md stays within scope (analyze adjacent clip boundary frames, validate handles, choose transition type, emit design instructions). It references the provided reference docs only. One operational ambiguity: it assumes access to '粗剪数据' and frame-level visual content (motion vectors, lighting, composition) but does not specify how those are supplied — this is an operational requirement, not a security mismatch.
Install Mechanism
okInstruction-only skill with no install spec and no code files. Nothing is written to disk or downloaded during install.
Credentials
okNo environment variables, credentials, or config paths are requested. The skill only needs clip metadata and frame analysis inputs, which are proportional to its purpose.
Persistence & Privilege
okalways is false and the skill does not request elevated or permanent presence or modify other skills. Autonomous invocation is allowed (platform default) and is not combined with other red flags.