wechat-topic-research

Search WeChat public account articles and save public article text for topic research.

1|Updated Jul 13, 2026
One-click install
npx skills add https://github.com/HalseyYang/Skills-HY --skill wechat-topic-research-halseyyang
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: wechat-topic-research
Source: https://github.com/HalseyYang/Skills-HY/tree/main/wechat-topic-research
Command: npx skills add https://github.com/HalseyYang/Skills-HY --skill wechat-topic-research-halseyyang

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires cheerio, and includes scripts (resource) components.

What problem does it solve? Finding writing references on WeChat is hard because search results are noisy, outdated, or duplicated, and article bodies are difficult to capture. This Skill searches WeChat public account articles by keyword, filters old and duplicate results, and saves the plain text of selected public articles so you can build topic angles from real source material. ## Core Features & Use Cases - Keyword Search with Filtering: Search via Sogou Weixin with a configurable date window (default 30 days) and result limit (default 10), with automatic deduplication and exclusion reasons recorded in result.json. - Single-Article Text Capture: Read one public mp.weixin.qq.com article at a time, saving article.md, article.json, and source.html, with verification that the body matches the requested URL. - Safe Failure Handling: Stops immediately on captcha, 403, or 429 responses without retrying, rotating identities, or reading cookies, and reports honest statuses like blocked, unresolved, or filtered_empty. - Use Case: Ask for recent articles about a product topic, pick up to 3 promising candidates, save their text, and receive 2-3 writing angles grounded in the actual article content with titles, accounts, dates, and links. ## Quick Start Use the wechat-topic-research skill in this workspace to search WeChat articles from the last 30 days about "home organization", pick one public article, save its text, and summarize the main points with the saved file path.

Frequently Asked Questions about wechat-topic-research

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I search WeChat public account articles by keyword?

Run the research script with the search command, a keyword, and an empty output directory, optionally setting --days and --limit. It queries Sogou Weixin, filters results by date and duplicates, and writes candidates plus exclusion reasons to result.json.

How can I save the text of a WeChat article locally?

Use the read command with a public mp.weixin.qq.com article URL and a new output directory. The script saves article.md, article.json, and source.html, then verifies the body matches the requested URL before reporting body_saved.

What Node.js version and dependencies does this tool need?

It requires Node.js 22 or later and a single npm dependency, cheerio 1.1.2, installed locally in the skill directory with npm ci. No Python or global installs are needed for the built-in article collection.

Why does WeChat article collection stop with a blocked status?

The script stops when it detects captcha pages, HTTP 403, or 429 responses, and it never retries, rotates proxies, or uses cookies. You can open the article in your own browser and then provide the public link for a single-article read.

Can it download images, videos, or an entire account's history?

No. It saves only the text of one public article per run and does not fetch images, videos, comments, or full account archives. Paid, deleted, image-only, or audio-video articles are unsupported.

What are the limitations of WeChat article search results?

Dates come from the search index and only the returned page is filtered, so empty results do not prove no recent articles exist. Search rankings also do not indicate reading volume or quality, and summaries alone should not be treated as article content.