wechat-article-forge_skill

This skill streamlines WeChat article creation from topic ideation to draft publication, ensuring quality gates and structured formatting.
  • Python

1.8k

GitHub Stars

5

Bundled Files

2 months ago

Catalog Refreshed

3 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill openclaw/skills --skill wechat-article-forge

  • _meta.json298 B
  • CHANGELOG.md5.5 KB
  • README.md11.4 KB
  • SKILL.md9.9 KB
  • skill.yml8.2 KB

Overview

This skill orchestrates end-to-end WeChat Official Account article production: topic research, Chinese-first drafting, automated review and fact-checking, illustration generation, formatting, and saving the final result as a WeChat draft. An orchestrator routes tasks to specialized subagents and enforces quality gates, while the user completes final publication. It prioritizes preserving article intent and strict review thresholds to avoid drift and low-quality artifacts.

How this skill works

The orchestrator manages a nine-step pipeline that persists state and routes work to writer, reviewer, fact-checker, and illustrator subagents. The writer produces Chinese-first drafts from verified sources; a blind reviewer scores craft dimensions; automated and human revisions happen as needed; fact-checker verifies claims; formatting and preview occur before images are generated and the draft is saved to the WeChat draft box using the preferred API path or browser automation fallbacks.

When to use it

  • Producing WeChat Official Account articles with strict quality and fact-check requirements
  • Automating repeatable article workflows while preserving a defined voice
  • Generating text-first drafts that require later human approval for final publish
  • Creating illustrated scrapbook-style articles where image placement must be precise
  • Saving consistent drafts to WeChat when you want reproducible, auditable pipeline state

Best practices

  • Always supply or verify wechat API credentials to use the reliable API path and avoid browser automation
  • Keep the article’s original purpose statement (初心) intact across revisions to prevent topic drift
  • Use the voice training command to align writer subagent output with past best-performing articles
  • Run preview and human approval before generating illustrations to avoid wasted image costs
  • Respect the reviewer rubric thresholds and allow the defined auto/human revision cycles before proceeding

Example use cases

  • Turn a trending topic into a polished WeChat tutorial with sourced references and reproducible pipeline state
  • Batch-produce opinion or news pieces with enforced craft scoring and mandatory fact-check cycles
  • Train a custom voice from existing articles and generate drafts that match that profile
  • Create illustrated science explainer articles with controlled image counts and scrapbook layout
  • Automate formatting and WeChat-draft saving while leaving final publish action to an editor

FAQ

The pipeline saves the article to the WeChat draft box; final publish is intentionally manual. If API credentials are present the skill uploads reliably, but user still triggers publish in the official account console.

How many automated revision cycles run before human intervention?

The writer subagent runs up to two automated revise cycles. If reviewer scores remain below threshold, the pipeline pauses for human direction.

How are images handled to avoid WeChat upload issues?

Images are generated only after text approval via a controlled scrapbook pipeline and inserted using a two-phase injection (text first, images via clipboard blob paste) to ensure correct positions and CDN upload.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational