Fetches WeChat article read, share, and like metrics and backfills history records with recommendations.
npx skills add https://github.com/oaker-io/wewrite --skill wewrite-stats
Please help me install this Agent Skill. Skill: wewrite-stats Source: https://github.com/oaker-io/wewrite/tree/main/skills/wewrite-stats Command: npx skills add https://github.com/oaker-io/wewrite --skill wewrite-stats
After publishing WeChat Official Account articles, creators lack a structured way to pull performance data and connect it back to their content decisions. This Skill retrieves article metrics via the WeChat API, backfills them into the local history record, and turns raw numbers into actionable adjustments for future topics, titles, and frameworks.
wewrite stats --days 7 to pull reads, shares, likes, and read rates from the WeChat data analysis API.history.yaml and writes the stats fields back for long-term tracking.Ask the AI to review how my recent WeChat articles performed over the last 7 days and suggest improvements for my next topics and titles.
High-intent search queries and answers about installing and using this skill.
Run the stats review flow, which executes `wewrite stats --days 7` to call the WeChat data analysis API. It pulls reads, shares, likes, and read rates for recent articles and matches them to your local history records.
After metrics are backfilled into history.yaml, the Skill compares best and worst performing articles and attributes differences to title strategy, topic heat, or framework type. It then recommends concrete adjustments for future topics, titles, and structures.
Yes, fetching live data requires WeChat API credentials configured in config.yaml. Without credentials, the Skill falls back to qualitative analysis based only on existing history.yaml records and tells you so explicitly.
WeChat article metrics need time to accumulate, so newly published articles should be reviewed after about 24 hours. The Skill will advise you to wait rather than report incomplete numbers.
Backfilled stats are read by the topic-selection module, which weights frameworks and enhancement strategies that historically performed well. This forms a data feedback loop alongside the edit-learning flywheel.
lihanghangGenerate HTML content, covers, and publish drafts via the WeChat API.
Marwan78888Generate 3-4 article topic directions with titles, outlines, and pros/cons analysis.
created-by-forgeCross-reference Obsidian qualitative signals with Supabase performance data.
jiyangnanPost articles and image-text messages to WeChat Official Account via API or Chrome CDP.