Xiaohongshu

提供小红书公开笔记和作者主页的关键词搜索、内容摘要及时序表现对比分析功能。

Security Scan

Instructions are consistent with a tool for summarizing public Xiaohongshu (小红书) pages and author metrics; there are no undeclared credentials or installs, but some guidance is vague around login and rate-limiting so exercise caution.

Purpose & Capability

Name/description (search/summarize public Xiaohongshu notes and author pages) matches the SKILL.md content. The skill asks for no binaries, env vars, or installs, which is proportionate for an instruction-only public-page analysis helper.

Instruction Scope

SKILL.md instructs the agent to access public search/author pages and extract specific fields, and it explicitly discourages bulk scraping and automated login. However some instructions are high-level (e.g., '分页解析', '请求频率需控制') and leave implementation discretion to the agent, which could lead to excessive requests or attempts to bypass anti-bot controls if not constrained.

Install Mechanism

No install spec and no code files — lowest-risk form. Nothing is downloaded or written to disk by the skill itself.

Credentials

The SKILL.md mentions '登录态仅用于个人页信息整理' (login state may be used for personal page info) but the skill declares no required credentials or config. This is a mild inconsistency: if the agent later asks for cookies/credentials to access private or account-bound pages, those requests would be outside the declared requirements and should be treated carefully.

Persistence & Privilege

always is false and there is no mechanism to persist or modify other skills or system-wide settings. The skill does not request elevated presence.

Assessment

This skill is coherent for analyzing public Xiaohongshu pages, but you should: (1) refuse to provide account credentials, session cookies, or secrets unless you explicitly trust the agent and understand the risk—SKILL.md mentions login state but declares no credentials; (2) monitor and limit the agent's scraping behavior (rate limits, request caps) to avoid ToS violations or triggering platform defenses; (3) be cautious if the agent later requests installation of headless browsers or scraping libraries (these are higher-risk installs); (4) prefer using official APIs or platform-provided endpoints when available; and (5) if you want stronger guarantees, ask the skill author to clarify how login-requiring data is handled and to add explicit safeguards around request frequency and data retention.

小红书

用途与边界

• 面向公开笔记与作者主页的检索与表现数据摘要
• 不提供下载、接口逆向或突破风控的能力
• 仅用于公开页面的轻量分析与提醒

关键入口

• 主页：https://www.xiaohongshu.com/
• 发现页：https://www.xiaohongshu.com/explore
• 搜索入口：https://www.xiaohongshu.com/search

常见任务

• 指定关键词/话题的笔记集合摘要（点赞/收藏/评论）
• 作者主页近N篇笔记的时序表现对比
• 类目分布与内容题材分析

数据字段

• 笔记ID、标题/封面文案、作者昵称与主页链接、话题标签
• 点赞/评论/收藏数、发布时间、笔记链接、封面URL
• 品类/主题标签、是否种草/测评倾向（可人工标注）

自动化要点

• 动态渲染与人机校验，建议人工打开后触发解析
• 请求频率需控制，避免风控；不进行批量抓取
• 登录态仅用于个人页信息整理，不自动执行交互

示例流程

• 关键词分析：输入关键词 → 搜索 → 分页解析 → 输出交互指标
• 作者对比：访问主页 → 抽取近N篇笔记 → 统计时序表现
• 题材统计：收集条目 → 基于标签/文案分类 → 生成分布报告

合规提示

• 遵守平台规则，不进行下载、接口调用或批量采集
• 结果仅用于内部分析与提醒