No red flags found in any of the 11 categories — no credential harvesting, no data exfiltration, no curl-pipe-shell installer. Scanned against the SlowMist agent-security taxonomy, refreshed every 8 hours. Full audit →
by yuwen-cool · Agent Tool · ★ 739
Last updated: · Indexed by AgentSkillsHub · Auto-synced every 8h
🔒 Is yuwen-publish-precheck safe to install? View the security audit →
yuwen-publish-precheck｜发布前审 发抖音、小红书、微信视频号之前，先让 AI 帮你审一遍：能不能发、哪句有问题、怎么改——并且把你踩过的坑记住，越用越准。 做内容的人都困在同一个死循环里：选题平淡没流量，选题劲爆就违规。而平台不会在发布前告诉你哪里会出问题，等它提醒你的时候，已经是违规通知了。 其实平台拦的大多是表达方式，不是话题本身——同一个劲爆选题，几乎总有合规的爆款在正常拿流量。这个 Skill 帮你在发布前找到"同样劲爆但不违规"的说法。 30 秒看懂它怎么工作 两层检测是它和"违禁词表"的本质区别：词面命中只是线索，违规与否由语境决定。"月入过万"写在个人经历里没事，绑着卖课链接就违规——词表工具分不出这个，它可以。 看一个真实例子 下面是一段典型的带货口播稿和它的实际输出（测试实录节选）： 原稿： 姐妹们，我挖到宝了！这款酵素梅我吃了两周瘦了八斤，全网最低价只在我这，官方渠道根本拿不到这个价。现在下单还送体验装，想要的加我微信 vx8899，手慢无！ 审核结论（节选）： text 结论：原稿不建议发（抖音、小红书都不行），改完 3 处必改后可以发。 【必改】 "吃了两周瘦了八斤" → 食品宣称减肥功效 + 期限绑定确定效果（M02 极高风险）。 这个卖点本身就不能这么说——哪怕你真的瘦了八斤。改字眼保留主张属于"换
| Stars | 739 |
| Forks | 71 |
| Language | Python |
| Category | Agent Tool |
| License | MIT |
| Quality Score | 64.1778620748475/100 |
| Last Updated | 2026-09-16 |
| Created | 2026-07-15 |
| Platforms | claude-code, python |
| Est. Tokens | ~9k |
These tools work well together with yuwen-publish-precheck for enhanced workflows:
Looking for a yuwen-publish-precheck alternative? If you're comparing yuwen-publish-precheck with other agent tool tools, these 6 projects are the closest alternatives on Agent Skills Hub — ranked by topic overlap, star count, and community traction.
Token conscious database MCP server for Postgres, MySQL, SQL Server, Oracle, MariaDB, SQLite.
Supercharge AI coding agents with portable skills. Install, translate & share skills across Claude Code, Curso
Universal Claude Code workflow plugin with agents, skills, hooks, and commands
Private web search for AI assistants via SearXNG — supports Claude, Cursor, and any MCP client
Official remote MCP server for Atlassian. Securely connect Jira, Confluence, Jira Service Management, Bitbucke
The missing linter and lsp for AI coding assistants. Validate CLAUDE.md, AGENTS.md, SKILL.md, hooks, MCP. Plug
Explore other popular agent tool tools:
yuwen-publish-precheck is 发布前审｜发抖音/小红书/视频号前先让 AI 审一遍：哪句踩线、依据哪条官方规则、给能直接用的改法。38 篇真实样本校准判定尺度，80+ 条官方原文引文可查证，你踩过的坑沉淀成本地规则库越用越准。不承诺过审，不教绕审。. It is categorized as a Agent Tool with 739 GitHub stars.
yuwen-publish-precheck is primarily written in Python. It covers topics such as agent-skills, ai, chinese.
You can find installation instructions and usage details in the yuwen-publish-precheck GitHub repository at github.com/yuwen-cool/yuwen-publish-precheck. The project has 739 stars and 71 forks, indicating an active community.
yuwen-publish-precheck is released under the MIT license, making it free to use and modify according to the license terms.
The top alternatives to yuwen-publish-precheck on Agent Skills Hub include dbhub, skillkit, claude-workflow-v2. Each offers a different approach to the same problem space — compare them side-by-side by stars, quality score, and community activity.
Grades come from a rule-based scan built on the SlowMist agent-security taxonomy, covering 11 red-flag categories including credential harvesting, data exfiltration, and curl | sh installers. It is a first-layer scan, not a manual audit — we say so rather than overstate it.
The scale of the problem is documented independently: Liu et al. (2026), in a study of 31,132 agent skills, report that 26.1% contain security vulnerabilities. Our own full-catalog census is published as a citable open dataset.
Sources & who's responsible: