Footer

Your go-to destination for discovering trending open-source projects and uncovering the insights that matter

Legal

  • Terms and conditions
  • Privacy policy

More

  • Pricing
  • Advertise
  • onedump
  • gosnakego
  • Contact us

Social

  • X (Twitter)
  • GitHub
Trendshift owl
What they build speaks louder than words
© 2026 Trendshift, created by Julian Li with ❤️
SignalAdvertise
Trendshift
SignalAdvertise
Sign in
Trendshift owl

Discover trends that matter

SignalAdvertise
DailyWeeklyMonthlyYearly
Live mentionsTopics
RepositoriesDevelopers
InsightsStats
Bifrost
Bifrost
Busbar
Busbar

larashero3-dotcom/lieflat-gongwen

通过102万字语料提炼的公文写作 skill，把公文写作风格变成可测量、可验收的量化数据

#AI skills
Visit GitHub
Python
2.5k
333
1 contributors
Custom license
AchievementsView all →
FeaturedAdvertise to 180K+ monthly visitors
Bifrost

Fastest Enterprise AI Gateway for Routing, Governing, and Securing all your AI Traffic

Busbar

Control what AI can do before it acts

Momentum

How fast larashero3-dotcom/lieflat-gongwen is gaining stars, and where it lands on Trendshift and GitHub Trending

Activity and trending

Stars, forks and merged pull requests each day or month, lined up with the repository's rank on Trendshift Trending

Sep 9 – Nov 8, 2026

Stars +2,282 in 61 days

0200400
374

Forks +315 in 61 days

0204060
59

Merged PRs 0 in 61 days

01

Trendshift Trendingwhere the repository ranked each day it made the top 25

Top 25 overall

21

Top 25 in Python

61920
Sep 9Sep 17Sep 25Oct 3Oct 11Oct 19Oct 27Nov 4
The left and right arrow keys step from day to day; Home and End go to the first and the last

Stars, forks, merged pull requests and Trendshift Trending rank, by day
DayStars gainedForks gainedMerged PRsAll languagesPython
Wed, Sep 9, 2026+374+590#21#6
Thu, Sep 10, 2026+188+400Not trendingNot trending
Fri, Sep 11, 2026+81+150Not trendingNot trending
Sat, Sep 12, 2026+48+60Not trendingNot trending
Sun, Sep 13, 2026+56+70Not trendingNot trending
Mon, Sep 14, 2026+51+50Not trendingNot trending
Tue, Sep 15, 2026+24+20Not trendingNot trending
Wed, Sep 16, 2026+16+20Not trendingNot trending
Thu, Sep 17, 2026+16+00Not trendingNot trending
Fri, Sep 18, 2026+10+10Not trendingNot trending
Sat, Sep 19, 2026+13+20Not trendingNot trending
Sun, Sep 20, 2026+22+50Not trendingNot trending
Mon, Sep 21, 2026+22+70Not trendingNot trending
Tue, Sep 22, 2026+26+40Not trendingNot trending
Wed, Sep 23, 2026+13+10Not trendingNot trending
Thu, Sep 24, 2026+16+20Not trendingNot trending
Fri, Sep 25, 2026+6+00Not trendingNot trending
Sat, Sep 26, 2026+8+30Not trendingNot trending
Sun, Sep 27, 2026+24+20Not trendingNot trending
Mon, Sep 28, 2026+25+00Not trendingNot trending
Tue, Sep 29, 2026+17+10Not trendingNot trending
Wed, Sep 30, 2026+10+20Not trendingNot trending
Thu, Oct 1, 2026+9+00Not trendingNot trending
Fri, Oct 2, 2026+6+20Not trendingNot trending
Sat, Oct 3, 2026+6+00Not trendingNot trending
Sun, Oct 4, 2026+8+00Not trendingNot trending
Mon, Oct 5, 2026+5+00Not trendingNot trending
Tue, Oct 6, 2026+15+10Not trendingNot trending
Wed, Oct 7, 2026+30+20Not trendingNot trending
Thu, Oct 8, 2026+53+20Not trendingNot trending
Fri, Oct 9, 2026+73+100Not trendingNot trending
Sat, Oct 10, 2026+13+30Not trendingNot trending
Sun, Oct 11, 2026+8+00Not trendingNot trending
Mon, Oct 12, 2026+14+10Not trendingNot trending
Tue, Oct 13, 2026+12+30Not trendingNot trending
Wed, Oct 14, 2026+8+10Not trendingNot trending
Thu, Oct 15, 2026+5+10Not trendingNot trending
Fri, Oct 16, 2026+8+20Not trendingNot trending
Sat, Oct 17, 2026+5+10Not trendingNot trending
Sun, Oct 18, 2026+5+10Not trendingNot trending
Mon, Oct 19, 2026+27+20Not trendingNot trending
Tue, Oct 20, 2026+15+20Not trendingNot trending
Wed, Oct 21, 2026+19+30Not trendingNot trending
Thu, Oct 22, 2026+14+00Not trendingNot trending
Fri, Oct 23, 2026+13+20Not trendingNot trending
Sat, Oct 24, 2026+12+30Not trendingNot trending
Sun, Oct 25, 2026+6+10Not trendingNot trending
Mon, Oct 26, 2026+7+20Not trendingNot trending
Tue, Oct 27, 2026+12+10Not trendingNot trending
Wed, Oct 28, 2026+7+10Not trendingNot trending
Thu, Oct 29, 2026+8+00Not trendingNot trending
Fri, Oct 30, 2026+11+30Not trendingNot trending
Sat, Oct 31, 2026+4+10Not trendingNot trending
Sun, Nov 1, 2026+280+430Not trending#19
Mon, Nov 2, 2026+280+300Not trending#20
Tue, Nov 3, 2026+64+90Not trendingNot trending
Wed, Nov 4, 2026+39+70Not trendingNot trending
Thu, Nov 5, 2026+42+50Not trendingNot trending
Fri, Nov 6, 2026+36+30Not trendingNot trending
Sat, Nov 7, 2026+28+10Not trendingNot trending
Sun, Nov 8, 2026+9+00Not trendingNot trending
Total+2,282+3150
  • 1#1 on the list
  • 7Rank that day
  • Not trending

Mentions and achievements

Where people are talking about larashero3-dotcom/lieflat-gongwen, and every badge it has earned
  1. Nov 21 post

    • 写材料的痛点是AI味太重。这项目拿102万字真实公文，把文风蒸馏成Agent Skill。我原以为又是套壳，结果它直接喂统计规律。仓库：

      Read full on X, posted by @yang3izc

      @yang3izc · x.com

  2. 27 days earlier
  3. Oct 61 post

    • 102 万字真实公文喂出来的写作方法，最后长成了这套「公文写作 DNA」Skill。 它挑的活儿很刁：专攻 7 种没固定模板、最难下笔的文体——调研报告、领导讲话、工作方案、经验材料这些；通知、请示之类有法定格式的，它不碰。 GitHub： 用起来步骤很清楚：判文种 → 调该文种的参数和范文 → 先出骨架停住等确认 → 按参数写正文 → 跑自检脚本，查文种有没有写错位。…

      Read full on X, posted by @shenxiankk

      @shenxiankk · x.com

  4. 8 days earlier
  5. Sep 281 post

    • • Covers seven types of official documents • Checks your draft against real writing patterns • Works with common AI writing assistants

      Read full on X, posted by @the_osps

      @the_osps · x.com

  6. Sep 251 post

    • lieflat-gongwen 这个开源项目，体制内写材料的朋友不妨重点瞅瞅。 普通 AI 写公文最劝退的地方就是“AI 味”扑面而来：套话成堆、结构齐整，可一眼就露馅，根本不像真材料。 这项目直接拿 102 万字真实公文做统计分析，把公文写作那套风格硬生生“蒸馏”成一个 Agent Skill。

      Read full on X, posted by @kfhshl

      @kfhshl · x.com

  7. 11 days earlier
  8. Sep 141 post

    • 「公文写作 DNA」这套 Skill 确实有料：作者用 102 万字真实公文提炼出一套可复用的写作方法。 它主攻 7 种“没固定模板、最难写”的文体：调研报告、领导讲话、工作方案、经验材料等；通知、请示这类有法定格式的不在范围内。 GitHub： 使用流程也很干脆：先判文种 → 读该文种参数和范文 → 先出骨架停下等确认 → 再按参数写正文 → 最后跑自检脚本，查文种是否写错位。…

      Read full on X, posted by @Jolyne_AI

      @Jolyne_AI · x.com

  9. Sep 121 post

    • 推荐一个开源项目 lieflat-gongwen，体制内写材料的可以重点看看。 一般的 AI 写公文，最大的问题就是“AI 味”太重：套话一堆、结构整齐，但一眼就不像真正的材料。 这个项目干脆拿 102 万字真实公文做统计，把公文写作风格直接“蒸馏”成一套 Agent Skill ⁠

      Read full on X, posted by @fhwofjow51260

      @fhwofjow51260 · x.com

  10. Sep 111 post

    • 「公文写作 DNA」Skill 已从102万字真实公文中提炼，覆盖调研报告、领导讲话、工作方案、经验材料等7类非固定模板文体。它的关键不在“公文”二字，而在边界清楚：通知、请示等法定格式不做。GitHub：

      Read full on X, posted by @Aji_Notes

      @Aji_Notes · x.com

  11. Sep 103 posts

    • 「公文写作 DNA」 这个 Skill 有点东西，作者拿了 102 万字真实公文提炼出来的公文写作技能。 覆盖调研报告、领导讲话、工作方案、经验材料等 7 种没有固定模板的文体，通知、请示这类有法定格式的它不做。 GitHub： 写之前先判文种、读该文种的参数和范文，出骨架停下等确认，再按参数写正文，最后跑自检脚本查文种有没有错位。 遇到拿不准的数字一律留「待补」占位，不编数。…

      Read full on X, posted by @GitHub_Daily

      @GitHub_Daily · x.com

    • lieflat-gongwen 是一个通过102万字语料提炼的公文写作 skill，将公文写作风格转化为可测量、可验收的量化参数（句长、顿号密度、标题结构等），供 Codex、Claude Code、Moxt 等兼容 agents.md 规范的 AI Agent 调用。

      Read full on X, posted by @weiwei2018831

      @weiwei2018831 · x.com

    • 公文写作最该摆脱的，可能不是“不会套模板”，而是把不同文种写成同一种腔调。 这个 skill 用102万字语料把句长、标题层级、并列结构等风格差异量化，还区分工作方案、调研报告、党建材料的文种指纹，并用脚本只拦截结构冲突。它的启发在于：先判文种再写，比先找万能模板更可靠。

      Read full on X, posted by @vintcessun

      @vintcessun · x.com

You might also like

Repositories close to larashero3-dotcom/lieflat-gongwen
FeaturedAdvertise to 180K+ monthly visitors
Bifrost

Fastest Enterprise AI Gateway for Routing, Governing, and Securing all your AI Traffic

Busbar

Control what AI can do before it acts

duyunzhoudyz/dyz-gongwen-skills

Python

公文写作 AI Agent Skills：懂规矩、懂单位、懂角色；不编造、守保密、帮你做减法。覆盖 22 种文书类型，支持 Claude Code、Codex、Cursor 等主流工具，防编造、懂国标格式、适配机关事业单位和医院高校国企。

45316

gongyu0918-debug/chinese-official-writing-skill

Python

中文公文与正式工作材料的起草、改写、压缩与复核，覆盖通知、请示、报告、函、方案、会议纪要、可研及 AI 算力等正式文本。

1019

zhaohui-yang/official-document-drafting

Python

A Codex skill for drafting, standardizing, and exporting Chinese official documents and formal reports to Markdown and Word (.docx).

14819

Imbad0202/tw-formal-writing

Python

台灣正式文件撰寫 AI Skill：公文、存證信函、陳情書、合約 — 匯入 ChatGPT / Claude / Gemini 即可使用

19729

linhut/gongwen-skill

Python

公文全流程处理工具——基于 GB/T 9704《党政机关公文格式》 国家标准，面向公文写作、企事业单位材料编制场景，支持 格式检查与修复、内容生成优化（Word 原生修订+批注/差异对比版）、模板生成、Markdown 转公文、版头版记页码注入、事实核验、风格增强 等完整能力。原生支持 DeepSeek Harness (DSH) 技能系统，打包为可被 AI Agent 直接调用的 Skill，完全自包含，克隆即用。

10517

KaguraNanaga/document-format-skills

Python

公文格式处理Skill: Professional Word document formatting toolkit for Chinese official documents

23920

linhut/document-ai-assistant

Python

AI 公文智能优化助手 — 基于 GB/T 9704 标准的公文格式智能检测与优化桌面应用，可本地离线部署

12223

luoling8192/technical-writing

中文内部技术写作的 agent skill，约束设计文档 / 评审稿 / postmortem / 分享稿场景的语气、句法、结构

22013

surendranb/writing-skills

Python

Procedural writing-style skills for AI agents: official frameworks (plain language, business writing, GOV.UK, AP style, STE-100, developer docs) + character voices. Agent Plugins 1.0.0, works in opencode, Claude Code, Codex.

533