Skip to main content
skill CX Codex

wechat-moments — Claude Code skill: what it does, how to install

Skill for Claude Code ranks and summarizes the user WeChat Moments feed. Posts are filtered by event weight and reply history with each author.

MIT 107 installs ★ 750 updated 2026-10-21

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓
npx skills add OpenClaudia/openclaudia-skills --skill wechat-moments -a claude-code
Open on GitHub

What wechat-moments does

The skill turns a WeChat Moments feed into a concise digest, filtering out promotion and small talk. Scoring combines event importance from major life milestones down to daily routines with relationship weight based on outbound messages.

A separate repository supplies the scripts and a pre-exported SQLite database. The client must stay open during export because sleep gaps cannot be recovered.

The tool never decrypts anything or handles keys directly. Automated desktop client interaction carries account risks if detected by Tencent.

What it helps with

  • summarize WeChat Moments feed
  • filter out marketing posts from social feed
  • rank contacts by message history
  • build a prioritized social media digest

Example prompts for the agent

  • summarize my Moments
  • what did I miss on Moments
  • check my WeChat Moments

How to install wechat-moments

1Skills CLI (Vercel)

npx skills add OpenClaudia/openclaudia-skills --skill wechat-moments -a claude-code

2Manually: put the files in place

~/.claude/skills/wechat-moments/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/wechat-moments/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about wechat-moments

How to install wechat-moments in Claude Code?

Run npx skills add OpenClaudia/openclaudia-skills --skill wechat-moments -a claude-code, then use ~/.claude/skills/wechat-moments/SKILL.md.

What does wechat-moments do?

It ranks and summarizes a WeChat Moments feed, separating real life updates from promotions based on author interaction history.

What is required to run wechat-moments?

A separate script repository, a running WeChat client, and a pre-exported plaintext database are required.

Source text

SKILL.md · MIT
---
name: wechat-moments
description: "Rank and summarize the user's own WeChat Moments (朋友圈) feed so real life events and genuine information rise above promotion and small talk, weighted by how much the user actually talks to each author. Use when the user says 'summarize my Moments', 'read my 朋友圈', 'what did I miss on Moments', 'rank my Moments', or 'check my WeChat Moments'."
---

# WeChat Moments digest

Turns a WeChat Moments feed into a short ranked digest: what deserves a reply,
what is genuinely informative, and what is marketing wearing a personal voice.

The scripts live in a **separate repository**, not here, because they depend on
the WeChat desktop client rather than any marketing API:

**https://github.com/OpenClaudia/wechat-moments**

```bash
git clone https://github.com/OpenClaudia/wechat-moments.git
```

Read that repo's README before running anything. It documents the setup, the
limits, and the risks — this skill is the ranking method, not the plumbing.

## Before you touch it

- **It does not decrypt anything and never handles keys.** It reads a plaintext
  database that a separate WeChat export tool has already produced. If the user
  has no export yet, that is step one and it is not this skill's job — the repo
  lists existing tools for it.
- **Moments do not sync in the background.** WeChat fetches the timeline only
  while the Moments window is open. Without a sync the export contains only
  what the user already scrolled past, so the digest is a replay of what they
  have already seen. Say so rather than presenting stale posts as today's.
- **Tell the user the limits before setting this up on a schedule.** It needs
  WeChat running and logged in, it only stays complete while the machine is
  awake (sleep gaps are permanently unrecoverable), Tencent may break or patch
  it at any time, automating the client may be noticed and may carry account
  risk, and nothing is guaranteed.
- **Never post, like, comment, or message on the user's behalf.** Reading and
  writing files is the whole remit.

## Ranking

This is the part worth carrying to any feed, not just WeChat.

**Score = event weight × relationship weight.** Do it yourself from the
structured records; do not write a scoring script. The point is judgement —
"her mother is in hospital" outranks "went to Kyoto".

**Event weight (0-10).** Birth, death, serious illness, wedding, divorce,
emigration at the top. New job, funding, graduation, a home, a pregnancy just
below. Travel and opinions in the middle. Food, gym, weather, scenery at the
bottom. Promotion and recruiting score zero.

**Insight vs. announcement** decides most of the digest, because feeds full of
founders and creators are full of engagement bait. Real information has at
least one of: a number you could check, a mechanism (*why* it works), a cost
the author actually paid, or a changed mind. Adjectives (突破 / 颠覆 /
groundbreaking), a milestone with no mechanism, an invitation, an upvote
request, or a takeaway generic enough to predate the event are announcements.
**Prestige is not evidence** — a famous venue is still promotion when the
payload is one generic sentence. A tiny post can be real insight.

**Relationship weight — lead with how much the user SENT.** Not total volume:
a marketer who broadcasts dozens of pitches and never gets a reply scores
identically to a close friend. A reply is a choice; an inbound blast is not.
Never-messaged contacts get filtered hard whatever they posted — expect them to
be about a third of any day's feed.

**A close friend's small post usually beats a stranger's big news**, because
only one of them is a relationship the user can act on. Ranking on content
alone inverts this every time.

**Time since last contact is its own signal.** Many messages sent but silent
for a long time is a lapsed close friend, and a life event from one of those is
the most actionable thing a digest can surface.

**Promotion wears a personal voice.** A discount code, a signup link, an
invite-only pitch: the poster is selling, however casual the phrasing.

**Urgency is a modifier, never a reason.** Never open a digest with a dated
section — that structure promotes whatever has a deadline, and deadlines are
what promotions have. An expiring ad is still an ad.

## Output

Ten items is plenty; fewer is often right, and **zero worth reporting is a
legitimate result** — say so rather than padding. Keep strong ties and
weak-tie-but-useful content in separate, clearly labelled sections, so
interesting content from a stranger never looks like something to act on. Show
the relationship numbers per entry so the ranking is auditable and the user can
correct it.

Summarize Chinese posts in Chinese and English posts in English. Say what a
post *is*; never repost it verbatim.

© OpenClaudia · MIT · OpenClaudia/openclaudia-skills

Similar skills

All skills for Agent setup: memory, context, orchestration →

From the same set73

The whole set →
github-stars github-stars renders terminal charts showing daily and hourly star growth for any GitHub repo in your chosen timezone. Track launches, Product Hunt spikes, or DevRel momentum in real time. CX Codex +1 197installs organize-skills organize-skills audits a skill library, applies names based on backend dependencies, and repairs links and references affected by renames. Useful when several tool hubs share the same library. CX Codex +1 139installs ai-citations-report ai-citations-report is a Claude Code skill that builds GEO reports for a domain. It checks Google AI Overview and ChatGPT prompts that cite the site and identifies content gaps. CX Codex +1 136installs geo-analysis geo-analysis runs a GEO/SEO analysis for a domain through the Enception API and downloads a PDF on AI visibility, the market and competitors. Handy for client audits. CX Codex +1 135installs geo-difficulty geo-difficulty scores how hard a keyword is to rank for by checking page-level authority of real competitors, AI Overview citations, and the site's current position. CX Codex +1 134installs marketing-slides marketing-slides writes the outline for a pitch or sales deck, then, once you approve it, generates the slides in ChatSlide. CX Codex +1 132installs task-banner task-banner displays a persistent macOS desktop banner when Claude Code finishes a task, with a one-line summary of what was done. Works even when Notification Center silently drops CLI notifications. CX Codex +1 126installs gsc-portfolio-audit gsc-portfolio-audit — a skill for Claude Code, Codex, and other platforms that audits every Google Search Console property in an account. It compares clicks, impressions, and keyword performance across periods. CX Codex +1 112installs

Project sponsors

Companies that keep this analytics open