QClaw
by @projectsnowwork
Summarize noisy WeChat group chats into decisions, action items, mentions, risks, and tracked keywords for OpenClaw workflows.
clawhub install qclaw
π About This Skill
name: QClaw
description: Summarize noisy WeChat group chats into decisions, action items, mentions, risks, and tracked keywords for OpenClaw workflows.
version: 1.0.0
metadata:
openclaw:
emoji: "π¬"
QClaw
QClaw is a WeChat group chat signal extraction skill.
Use this skill when the user wants to:
summarize a WeChat group chat
extract decisions and action items
track keywords such as project names, clients, or deadlines
identify direct or implicit mentions that may need attention
detect conflict, urgency, or risk in busy group discussionsWhat this skill does
compresses long group conversations into short, useful summaries
separates signal from noise
extracts decisions, tasks, blockers, and next steps
highlights mentions of tracked keywords
identifies messages that likely require the user's attentionHow to use
Prerequisites
OpenClaw installed
A WeChat group chat log pasted as text, or exported as txt/json
Optional tracked keywords supplied by the userAccepted input
plain text chat log
JSON array of messages with timestamp, sender, and content
copied chat transcriptBest output structure
1. Executive summary
2. Decisions made
3. Action items
4. Mentions and keyword hits
5. Risks / conflicts / urgency
6. Suggested follow-up
Privacy & Security
Prefer anonymized or redacted chat logs when possible
Do not retain chat content longer than needed for the current workflow
Mark uncertainty clearly
Never fabricate missing contextWhen NOT to use QClaw
personal 1-on-1 chats
groups with very few messages
real-time alerting workflows
legal or compliance workflows requiring verbatim records
βοΈ Configuration
OpenClaw installed
A WeChat group chat log pasted as text, or exported as txt/json
Optional tracked keywords supplied by the userAccepted input
plain text chat log
JSON array of messages with timestamp, sender, and content
copied chat transcriptBest output structure
1. Executive summary
2. Decisions made
3. Action items
4. Mentions and keyword hits
5. Risks / conflicts / urgency
6. Suggested follow-up
