XiaoHongShu

帮助生成和优化符合小红书风格的笔记内容，包括标题、话题标签和emoji推荐，提升内容吸引力与互动率。

Security Scan

The skill’s code and runtime instructions match its stated purpose (generating and polishing Xiaohongshu notes); there are minor implementation quirks (an unused AI SDK dependency and no declared credentials) but no signs of data-exfiltration or scope creep.

Purpose & Capability

The name/description (XiaoHongShu content creation) align with the files and SKILL.md: functions are text optimization, title/hashtag generation, emoji insertion and simple generation/analysis. The included code implements exactly those behaviors. The presence of a dependency on '@ai-sdk/google' is somewhat unexpected given SKILL.md does not mention external SDK usage, but the code imports the SDK and does not actually call it, so this looks like a minor implementation artifact rather than a capability mismatch.

Instruction Scope

SKILL.md describes inputs/outputs and runtime behavior only around content generation and optimization. The instructions do not ask the agent to read unrelated system files, environment variables, or send data to external endpoints. The provided index.js performs purely local string manipulation and simple heuristics; it does not contain network calls or file reads/writes beyond normal CLI usage.

Install Mechanism

There is no explicit install spec in the skill bundle (instruction-only), but package.json lists a dependency on '@ai-sdk/google'. Installing the skill in an environment that runs npm install would fetch that package from the registry. That is a standard dependency pattern, not inherently suspicious, but it does mean an install step could pull third-party code from the npm registry.

Credentials

The skill declares no required environment variables or credentials, and the SKILL.md does not request secrets. However, package.json depends on '@ai-sdk/google', which in other contexts might require API credentials to call Google services. In this code base the imported AI object is not used, so there is no active credential use or exfiltration in the present files — still, the dependency implies a potential need for credentials if the code were extended.

Persistence & Privilege

The skill does not request always: true, does not modify other skills or system-wide configuration, and does not persist privileged settings. It runs as a normal user-space CLI script and does not request elevated persistence or cross-skill privileges.

Assessment

This skill appears coherent and focused on Xiaohongshu-style content generation. Before installing, consider: 1) npm dependency: package.json depends on '@ai-sdk/google' — installing the skill may fetch that package from the npm registry and that package might contact external services or require API keys; verify its origin (inspect the package or the repo). 2) Source provenance: the skill's source is listed as unknown; if you need stronger assurance, review the GitHub repo linked in package.json or run the code in a sandbox. 3) Runtime model use: the SKILL.md doesn’t ask for credentials, and the included code doesn’t actually call external AI APIs, but if you (or an update) modify it to use the AI SDK, API keys could be required — ensure you don’t provide unrelated credentials. If you want to proceed safely: inspect the dependency '@ai-sdk/google', run npm install in an isolated environment, or request the author/source verification.

小红书内容创作技能 (xiaoHongShu)

简介

这是一个专门帮助 AI 生成和优化小红书平台内容的手工技能。

功能

1. 内容润色：优化内容，使其更符合小红书风格
2. 标题优化：生成吸引点击的小红书标题
3. 话题标签：自动生成相关话题标签
4. emoji 优化：为内容添加合适的 emoji
5. 内容生成：基于主题生成不同类型的小红书笔记

使用场景

• 为 AI 生成小红书笔记内容
• 优化文案使其更符合平台风格
• 生成吸睛的标题
• 推荐相关话题标签

输入

• text: 需要优化的内容文本
• topic: 内容主题（如：生活、美妆、美食、穿搭等）
• format: 内容格式（图文、视频脚本、纯文等）

输出

• 优化后的小红书笔记
• 多个标题选项
• 话题标签列表
• 风格分析报告

技能版本

v1.0