AI Agent Hub
Back to skills
Official Document Authoring Assistant icon

Official Document Authoring Assistant

Content Creation Updated 2026.08.30

Paste the following prompt into your AI chat to install this skill:

Please follow https://skillhub.cn/install/skillhub.md to install @user_a601cd18/official-document-skill

About this skill

The Problem with Chinese Official Documents

Writing Chinese official documents involves multiple pain points: confusion between document types (e.g., 请示 vs 报告, 批复 vs 复函), complex formatting rules (title structure, recipient addressing, closing formulas), and obvious AI artifacts (empty parallelism, inflated meaning, vague subjects). These issues result in files that are either structurally incorrect or sound machine-generated, making them unusable as deliverables.

How This Skill Works

Core Capabilities

The skill integrates three knowledge bases:

  • Formal document rules: Covers 12 statutory document types including 通知 (notice), 请示 (request for instructions), and (correspondence), with scenario-specific structures
  • Journalistic expression patterns: High fact density, restrained judgment, functional paragraphs, precise word choice
  • De-AI processing rules: Identifies and removes common AI tells like meaning inflation, fake depth, vague actors, and rule-of-three packaging

Key Steps

  1. Scenario identification: Determines issuer, recipient, writing relationship (upward/downward/parallel), purpose, and audience
  2. Document type matching: Auto-selects the correct type based on purpose and relationship; silently corrects user errors or flags mismatches
  3. Fact extraction: Gathers subject, object, actions, data, deadlines, responsibilities before drafting
  4. Skeleton construction: Builds structure according to document-type standards without forcing generic templates
  5. Draft generation: Prioritizes facts and tasks over judgment; uses long sentences for background and mechanisms, short sentences for decisions and reminders
  6. Dual validation: Runs anti-AI detection (fact density, judgment strength, sentence rhythm) and humanization pass (removes machine-like patterns)

Boundaries and Caveats

  • Input dependency: Requires real facts from the user; will not fabricate laws, documents, meetings, leader names, or data
  • Format simulation limits: Cannot simulate physical elements like seals or document numbers; generates text content only
  • Document coverage: Supports 12 formal types (通知, 请示, 报告, 批复, , 纪要, etc.) and practical formats (工作方案, 调研报告, 宣传稿, etc.)
  • Language default: Outputs in Chinese by default; users can specify other languages
  • Completeness principle: Delivers complete usable drafts when facts are sufficient; uses placeholders like 〔发文机关〕 with supplement lists when key information is missing

Use Cases

  • When requesting approval for special funds or projects from superiors, drafting a properly formatted one-matter-per-document request for instructions
  • After organizing special inspections, drafting a circular to commend advanced models, criticize violations, and propose rectification requirements
  • Preparing briefing materials for department annual summary meetings, organizing work achievements, existing problems, and next steps
  • After receiving public petition regarding policy implementation, drafting a petition response stating investigation results and relief channels

Best For

  • Office secretarial staff who regularly draft downward documents like notices and circulars
  • Policy research or general department positions responsible for writing research reports and work plans
  • Grassroots unit leaders or supervising officials who need to prepare speech drafts and stance statements
  • Petition handling staff who receive public letters and complaints and must provide formal written responses