Remove Defensive Writing in Chinese Official Documents (gongwen-direct)
Rewrites existing Chinese government documents to strip disclaimers, hedging and fence-sitting, turning soft 'suggestions' into properly tiered directives.
What it does
Takes an existing draft (often AI-generated) and fixes seven classes of defensive writing:
- Deletes disclaimers and boilerplate reminders such as "for reference only"
- Clears hedges ("possibly", "to a certain extent") and both-sides balancing sentences
- In deployment documents, converts "we suggest" / "you may consider" into tiered wording (must / shall / should / please) based on whether the recipient is a subordinate unit, a non-subordinate body, or the public
- Names vague actors, removes empty posturing, deletes AI-style over-questioning and meta commentary
- Adds the required stance elements: a recommendation in a 請示, an explicit yes/no in a 批復, and moves requests smuggled into reports out of the body
Two guardrails hold throughout: genuine risk and safety warnings are preserved, and the skill never decides on the user's behalf — missing stances become 【待补】 placeholders. With Python available it also runs a confidentiality check, a defensive-language lint, and a fact-diff between original and revision.
Who it's for
- Staff who draft or review Chinese official documents (notices, reports, requests, replies)
- Anyone whose AI drafts get criticized as "too soft, sounds like negotiating with subordinates"
- Document owners who need consistent certainty requirements per document type
Example uses
- A notice ending with "this plan is for reference only" — the line is deleted and "units are suggested to consider" becomes "units shall…".
- An ambiguous official reply — the skill inserts 【待补: approve/disapprove】 and flags exactly where a decision is owed.
- A report containing a budget request — the request is lifted out into the drafting notes, the closing becomes "hereby reported", and a separate request document is proposed.
· · · Install guide · · ·
Try it now, no install
Paste this into Claude to use the skill without installing anything.
Read the instructions in this file and follow them to help me: https://raw.githubusercontent.com/duyunzhoudyz/dyz-gongwen-skills/HEAD/skills/gongwen-direct/SKILL.md What I want: (describe your task here)
If Claude can't open the link, open it yourself and paste the contents instead.
↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.
Install in the Claude app (no terminal)
- Download the ZIP with the button below.
- In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
- Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
Install in Claude Code
Let Claude do it — paste this into Claude Code
Install the skill I found on AI Skill Mart. Copy the skills/gongwen-direct folder from the GitHub repo duyunzhoudyz/dyz-gongwen-skills into my ~/.claude/skills/gongwen-direct/. When it's done, tell me in one line what this skill can do.
Install with a command instead
git clone https://github.com/duyunzhoudyz/dyz-gongwen-skills.git && mkdir -p ~/.claude/skills && cp -r dyz-gongwen-skills/skills/gongwen-direct ~/.claude/skills/⚠ This is a third-party skill. Check the source repository before installing.
- Open a terminal.
- Clone the repository:
git clone https://github.com/duyunzhoudyz/dyz-gongwen-skills.git - Create the skills folder:
mkdir -p ~/.claude/skills - Copy the skill:
cp -r dyz-gongwen-skills/skills/gongwen-direct ~/.claude/skills/ - Verify that
references/_shared/and thescripts/folder came along; if the repo keeps shared rules elsewhere, copying the whole repo into~/.claude/skills/is safer. - Restart Claude Code, then paste your draft with a request like "remove the for-reference-only line and make this notice directive".
- Install Python 3 if you want the lint and fact-diff scripts to run; otherwise the skill falls back to a manual checklist.