Auto Redbook Content
Security checks across static analysis, malware telemetry, and agentic risk
Overview
This skill is coherent and low-risk: it generates Xiaohongshu-style rewrite prompts and saves JSON locally, with only purpose-aligned notes about optional MCP network fetching and local output files.
This looks safe to install if you want a local Xiaohongshu content-fetching and rewrite-prompt helper. Be aware that it may rely on a Xiaohongshu MCP tool for real network fetching, and it saves fetched note content plus rewrite prompts to local output files.
Static analysis
No static analysis findings were reported for this release.
VirusTotal
VirusTotal findings are pending for this skill version.
Risk analysis
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.
Using the skill may cause the agent to call an external Xiaohongshu MCP tool to retrieve public content.
The skill discloses that the agent environment may use a Xiaohongshu MCP tool to fetch network content. This is aligned with the skill purpose, but it depends on the external MCP tool being trusted and appropriately scoped.
通过 xiaohongshu MCP 获取首页热门笔记 ... 可能通过 xiaohongshu MCP 工具进行网络抓取
Install only if you intend to allow Xiaohongshu fetching, and verify that the MCP provider and platform usage are acceptable for your environment.
If fetched content contains prompt-like instructions, a later model call could be confused unless the content is handled strictly as source material.
Externally fetched note content is embedded directly into a prompt for later rewriting. This is expected for the skill, but untrusted social-media text could contain instructions that a downstream model should not treat as authoritative.
原标题：${note.original_title}\n原内容：${note.original_content}When using generated prompts, keep fetched note text clearly delimited as untrusted source content and do not let it override the user’s task.
Fetched posts and generated prompts may remain on disk after the skill runs.
The skill intentionally stores fetched content and generated rewrite prompts in local JSON files. This is disclosed and scoped to the output directory, but users should be aware that retrieved content persists locally.
本地存储：输出到 `output/xiaohongshu_YYYYMMDD_HHMMSS.json` ... JSON 文件包含：原始标题、内容、作者、去 AI 味改写提示词
Review or delete output files when no longer needed, especially if future runs fetch non-public or sensitive material through an MCP tool.