S
Solvync/humanizer
Writing rules and a linter for plain business writing, packaged as a skill for Claude.
更新 2026-11-15
在 GitHub 查看 ↗链接已于 2026-11-17 核查 · HTTP 200 · 打开的是厂商自己的官网
- ★ 数取值
- 0 ★ · Solvync/humanizer · 取自 GitHub 搜索 API · (2026-11-19 20:00:48 核对)
- 来源明细
- 从该仓库的 GitHub 页面直接读取 · 条目类型:开源项目
- 收录与活跃
- 收录于 2026-11-16 · 最近见到 2026-11-20 · 共见到 5 次 · 被 2 个来源列出
- 评分构成
- 综合分 203 · 2 个来源 · repo
常见问题(答案都来自我们的记录)
它适合谁 / 用来做什么?
我们把它归在这些标签下: 开源项目。标签来自条目自身,不是我们强加的分类。
谁维护它、还在更新吗?
仓库
Solvync/humanizer;我们收录于 2026-11-16;最近见到 2026-11-20。这里没有"官方认证"或"质量评级"——我们只记录来源与时间。README 原文(英文,未翻译) · 在 GitHub 读完整 README →
# Humanizer
以下摘录自该仓库的 README(片段)
Writing rules for text other people read, packaged as a skill for Claude. It covers email,
proposals, statements of work, reports, decks, captions, scripts and website copy.
The skill gives Claude four bans, twelve writing steps and a check to run before anything
ships. A Python linter enforces the mechanical half. Every rule ID it reports is defined in
references/tells.md, with specimens.
## The four bans
1. No em dashes, en dashes or spaced hyphens standing in for one.
2. No negative parallelism: a thing is never defined by pushing off its opposite.
3. Straight quotes only.
4. No emoji in professional copy.
## Install
### Claude apps
1. Download humanizer.zip from the latest release.
2. In Settings > Capabilities, turn on "Code execution and file creation". Newer layouts call
it "Create and edit files". On Team and Enterprise plans, an admin enables it in
organization settings.
3. Open Customize > Skills. Click "+", then "+ Create skill", then "Upload a skill".
4. Upload the ZIP. The skill appears in your skills list, where you can switch it on or off.
### Claude Code
bash
git clone https://github.com/Solvync/humanizer.git ~/.claude/skills/humanizer
Get updates later with:
bash
git -C ~/.claude/skills/humanizer pull
For one project only, clone it into that project's .claude/skills/humanizer folder.
### Running the linter directly
The scripts need Python 3.8 or newer and nothing else.
bash
python3 scripts/humanizer_lint.py draft.txt --profile prose --surface email
python3 scripts/humanizer_lint.py draft.txt --fix
python3 scripts/factcheck.py draft.md --facts facts.md
Exit codes: 0 pass, 1 flags only, 2 hard failure, 3 error. A flag is a question for a
person to decide.
## Adding your names and voice
Put the product and brand names you write about in PRODUCT_NAMES in
scripts/humanizer_lint.py. The linter then checks them like any other software name.
To write as a named person, add references/voice.md with that person's habits written as
instructions. SKILL.md lists what goes in it.
## Contributing
Issues and pull requests are welcome. Every change keeps these three checks passing:
bash
proposals, statements of work, reports, decks, captions, scripts and website copy.
The skill gives Claude four bans, twelve writing steps and a check to run before anything
ships. A Python linter enforces the mechanical half. Every rule ID it reports is defined in
references/tells.md, with specimens.
## The four bans
1. No em dashes, en dashes or spaced hyphens standing in for one.
2. No negative parallelism: a thing is never defined by pushing off its opposite.
3. Straight quotes only.
4. No emoji in professional copy.
## Install
### Claude apps
1. Download humanizer.zip from the latest release.
2. In Settings > Capabilities, turn on "Code execution and file creation". Newer layouts call
it "Create and edit files". On Team and Enterprise plans, an admin enables it in
organization settings.
3. Open Customize > Skills. Click "+", then "+ Create skill", then "Upload a skill".
4. Upload the ZIP. The skill appears in your skills list, where you can switch it on or off.
### Claude Code
bash
git clone https://github.com/Solvync/humanizer.git ~/.claude/skills/humanizer
Get updates later with:
bash
git -C ~/.claude/skills/humanizer pull
For one project only, clone it into that project's .claude/skills/humanizer folder.
### Running the linter directly
The scripts need Python 3.8 or newer and nothing else.
bash
python3 scripts/humanizer_lint.py draft.txt --profile prose --surface email
python3 scripts/humanizer_lint.py draft.txt --fix
python3 scripts/factcheck.py draft.md --facts facts.md
Exit codes: 0 pass, 1 flags only, 2 hard failure, 3 error. A flag is a question for a
person to decide.
## Adding your names and voice
Put the product and brand names you write about in PRODUCT_NAMES in
scripts/humanizer_lint.py. The linter then checks them like any other software name.
To write as a named person, add references/voice.md with that person's habits written as
instructions. SKILL.md lists what goes in it.
## Contributing
Issues and pull requests are welcome. Every change keeps these three checks passing:
bash
收录徽章(可选)
想把收录标记放进 README 就复制这行,可选,不影响收录。
[](https://anclum.com/zh/s/solvync-humanizer/)
- 来源
- github · https://github.com/Solvync/humanizer
- 按领域浏览
- 开源项目
- 收录来源
- 2 个独立来源
- 收录于
- 今天
更多条目
所属领域与标签
每个标签都能点进去看同类的全部条目。数字是当前库里的真实条数。