research-to-wechat

# Research to WeChat

Use this skill as a research-first control plane. Do not duplicate downstream skill wording.

## Core Rules

- Match the user's language.
- Ask one question at a time.
- Ask only when the answer changes source interpretation, structural frame, style fidelity, or draft publishing behavior.
- Keep Markdown as the canonical article asset until the HTML handoff.
- Save a draft only. Never publish live.
- Separate verified fact, working inference, and open question.
- Every major claim must be traceable to a source. Collect source URLs during research.
- Every article must end with a "## 参考链接" or "## References" section listing all sources.
- Apply the full normalization checklist to every article before refinement. Source artifacts, broken formatting, and LaTeX fragments must not survive into the final draft.
- Every inline image must pass a two-tier evaluation: first eliminate disqualifying defects, then verify content match to the surrounding text.
- Never pretend the workflow did interviews, long field research, team debate, or hands-on testing when it did not.
- Prefer visible disclosure of AI assistance and source scope. Refuse human-only framing that would misrepresent the process.
- Treat source capture as a runtime boundary: preserve title, author, description, body text, and image list before rewriting.

## Operating Paths

Route the request into one of two paths:

- `Path A: research-first article`
  use for: topic, keyword, question, notes, transcript, subtitle file
  goal: build the article from a research brief and evidence ledger

- `Path B: source-to-WeChat edition`
  use for: article text, markdown file, article URL, WeChat URL
  goal: preserve the useful source core, then rebuild it for WeChat reading and distribution

Default routing:
- procedural or tool-teaching material -> `tutorial`
- thesis, trend, strategy, critique, case material -> `deep-analysis`
- multi-topic roundup -> `newsletter`

## Capability Aliases

Resolve ca...

# Research to WeChat

Turn a topic, notes, article, URL, or transcript into a research-backed WeChat article with an evidence ledger, routed structure, polished Markdown, inline visuals, cover art, HTML output, a saved browser draft, and optional multi-platform distribution (小红书、即刻、播客、朋友圈).

## Install

**OpenClaw / ClawHub:**
```bash
clawhub install research-to-wechat
```

**Manual:**
```bash
curl -fsSL https://raw.githubusercontent.com/Fei2-Labs/skill-genie/main/research-to-wechat/scripts/install-openclaw.sh | bash
```

**Optional dependencies:**
- `md2wechat` skill — for HTML rendering (`clawhub install md2wechat` or [install guide](https://github.com/geekjourneyx/md2wechat-skill))
- Pencil MCP server — for article design templates from `design.pen`
- `WECHAT_APPID` / `WECHAT_SECRET` — for draft upload (not needed for research and writing)

## What this skill is

`research-to-wechat` is a research-first control-plane skill for WeChat article production.
It keeps one stable article contract from source packet to WeChat draft, while routing each stage to the right worker.
The key difference from a simple formatter is that this skill treats research assets, evidence limits, and structure choice as first-class outputs.

## Who this is for

- Writers who start from a topic and want a complete “research to draft” workflow
- Operators who need a repeatable WeChat article pipeline with traceable evidence
- Creators who publish to WeChat Official Accounts and optionally cross-post to other platforms
- Agents that need one consistent output contract across different source types

## What it produces

For every run, the skill creates one workspace:
`research-to-wechat/YYYY-MM-DD-<slug>/`

Expected files:
- `source.md`
- `brief.md`
- `research.md`
- `article.md`
- `article-formatted.md`
- `article.html`
- `manifest.json`
- `imgs/cover.png`
- inline image files referenced by the article body

The final Markdown must include these frontmatter keys:

- `title`
- `a...