← 返回 Skills 市场
3224
总下载
7
收藏
21
当前安装
4
版本数
在 OpenClaw 中安装
/install xiaohongshu-search-summarizer
功能描述
Searches Xiaohongshu(小红书) for a given keyword, extracts the top N posts (including texts, images, and user comments), and then synthesizes a comprehensive fi...
安全使用建议
This skill appears to do what it claims, but review these before installing: (1) Ensure you have playwright-cli, Python 3, and the Python 'requests' package installed and that you run the scraper in a safe, sandboxed environment — the script runs a headed browser and executes page JavaScript. (2) The scraper may require you to manually log into Xiaohongshu in the Playwright browser if the site challenges you. (3) Downloaded images and scraped text are written to the chosen OUTPUT_DIRECTORY and may contain personal data or copyrighted images — confirm you have the right to collect/process them and consider data retention policies. (4) The registry metadata omits declared binary requirements (minor inconsistency); verify prerequisites before use. If you need higher assurance, test the scripts in an isolated VM and inspect outputs before feeding local files to any automated agent or external service.
功能分析
Type: OpenClaw Skill
Name: xiaohongshu-search-summarizer
Version: 1.0.3
The skill provides a legitimate workflow for scraping and summarizing Xiaohongshu content using Playwright and Python. It includes security best practices such as domain whitelisting for image downloads in `scripts/parse.py` (restricting to `.xhscdn.com` and `.xiaohongshu.com`) and uses `ast.literal_eval` for safe parsing of stringified JSON. The instructions in `SKILL.md` are task-oriented and do not contain malicious prompt injections or attempts to exfiltrate sensitive data.
能力评估
Purpose & Capability
The skill's name and description match the included scripts: a Playwright-based scraper (run.sh) and a Python parser (parse.py) that downloads images and writes a raw markdown; however the registry metadata lists no required binaries while SKILL.md explicitly requires playwright-cli, python3, and the Python requests package — a minor metadata omission but otherwise coherent for the stated purpose.
Instruction Scope
SKILL.md limits agent actions to running the provided run.sh, reading the produced markdown and local image files, using vision capabilities to interpret images, and writing a synthesized markdown to the output directory. These instructions are within scope for a multi-modal summarizer and do not ask the agent to read unrelated system files or external secrets.
Install Mechanism
No install spec is included (instruction-only), which reduces risk. The skill depends on external tooling (playwright-cli, python3, pip package 'requests') that are expected to be preinstalled by the user; this reliance is reasonable but should be noted because the skill will fail if those are missing.
Credentials
The skill does not request any environment variables, credentials, or config paths. The runtime scripts only use non-sensitive environment variables for keyword and post limits, and the parser enforces image domain and HTTPS checks and disables redirects when downloading images, which is proportionate.
Persistence & Privilege
The skill is not always-enabled and is user-invocable only. It creates a temporary JS file in /tmp, runs Playwright, writes outputs into the user-specified output directory, and cleans up the temp script — it does not modify other skills or system-wide configuration.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install xiaohongshu-search-summarizer - 安装完成后，直接呼叫该 Skill 的名称或使用
/xiaohongshu-search-summarizer触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v1.0.3
- Added an explicit compatibility section specifying the required dependencies, including `playwright-cli`, `python3`, and the Python `requests` package.
- Updated the dependencies list in the documentation to mention that the `requests` package must be installed for image downloading.
- No changes to feature set or workflows; documentation improvements only.
v1.0.2
- Minor internal script adjustments, improving code organization and maintainability.
- No changes to usage, features, or user-facing documentation.
v1.0.1
- Adjusted the example script path in usage instructions for clarity and accuracy.
- No major logic or functionality changes; documentation improved for easier setup and more precise command references.
v1.0.0
- Initial release of Xiaohongshu Search and Summarize skill.
- Automates search, extraction, and synthesis of top Xiaohongshu posts (text, images, and user comments) into a comprehensive analytical report.
- Utilizes a headed browser via `playwright-cli` to bypass anti-scraping measures and extract high-quality multi-modal data.
- Reads both text and image files from collected data to generate a detailed, theme-driven summary, integrating visual analysis from post images.
- Requires users to run a shell script for raw data extraction and mandates file and image reading for in-depth analysis.
- Produces a polished final Markdown report grouping insights, themes, and nuanced perspectives, saved alongside raw data for easy access.
元数据
常见问题
Xiaohongshu Search Summarizer 是什么？
Searches Xiaohongshu(小红书) for a given keyword, extracts the top N posts (including texts, images, and user comments), and then synthesizes a comprehensive fi... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 3224 次。
如何安装 Xiaohongshu Search Summarizer？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install xiaohongshu-search-summarizer」即可一键安装，无需额外配置。
Xiaohongshu Search Summarizer 是免费的吗？
是的，Xiaohongshu Search Summarizer 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Xiaohongshu Search Summarizer 支持哪些平台？
Xiaohongshu Search Summarizer 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 Xiaohongshu Search Summarizer？
由 piekill（@piekill）开发并维护，当前版本 v1.0.3。
推荐 Skills