Story Review
AdvisoryAudited by Static analysis on May 12, 2026.
Overview
No suspicious patterns detected.
Findings (0)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
If you do not specify a narrow review target, the agent may inspect recently changed manuscript content and related story-planning files in the project.
The skill may read related project files and perform grep-style scans over the working directory when reviewing consistency. This is purpose-aligned for story review, but users should know the review may include project context beyond the exact draft text.
用户未指定 → 审查最近修改的内容（git diff）或当前章节 ... 读取相关的设定文件和大纲 ... 项目路径：{工作目录路径，用于 grep 扫描}Specify the chapter or file to review when scope matters, and use solo or lean mode if you want fewer tools/subagents involved.
Your draft text and project context may be shared with multiple reviewer subagents during full mode.
Full mode intentionally passes self-contained review context, including file paths and relevant content, to multiple named subagents. This is disclosed and central to the skill, but the subagents themselves are external to the provided artifacts.
使用 Agent 工具并行调用 4 次（不同 subagent_type）。每个 Agent 不继承父对话上下文，prompt 必须自包含文件路径和上下文。
Use full mode only when you trust the installed subagents; use solo mode for a simpler single-agent review of sensitive manuscripts.