← 返回 Skills 市场
Content System Humanizer Zh
作者
Abigale-cyber
· GitHub ↗
· v1.0.0
· MIT-0
192
总下载
0
收藏
1
当前安装
1
版本数
在 OpenClaw 中安装
/install content-system-humanizer-zh
功能描述
Remove obvious AI-writing traces from Chinese text in a constrained way. Use when Codex needs to reduce AI smell without changing facts, data, or the article...
使用说明 (SKILL.md)
Humanizer ZH
This repo-local skill is inspired by op7418/Humanizer-zh and blader/humanizer, but implemented for the content-system runtime.
Quick Start
.venv/bin/python -m skill_runtime.cli run-skill humanizer-zh \
--input content-production/drafts/harness-engineering-一人公司-article.md
What It Does
- Detects obvious AI-writing traces in Chinese prose
- Applies a surgical rewrite pass on body paragraphs
- Preserves facts, numbers, headings, structure, and article intent
- Writes a humanized copy plus a JSON report
Output
content-production/drafts/\x3Cslug>-humanized.mdcontent-production/drafts/\x3Cslug>-humanizer-report.json
Constraints
- Do not invent new facts or sources
- Do not flatten strong viewpoints into neutral filler
- Do not rewrite headings or titles by default
- Do not pretend the text comes from personal experience if it does not
安全使用建议
This skill appears internally consistent and does not request credentials or network installs. Before using it: (1) run it on a copy of your draft (it will write files under content-production/drafts), (2) review the generated <slug>-humanizer-report.json and the humanized output before publishing, and (3) confirm that the hosting environment provides the skill_runtime.writing_core implementation (the heavy lifting is delegated to that module — inspect or trust that library). Also note README references requirements.txt which is not included; ensure any needed Python dependencies are available in your runtime. If you want additional assurance, run the skill in an isolated workspace or sandbox first.
功能分析
Type: OpenClaw Skill
Name: content-system-humanizer-zh
Version: 1.0.0
The skill is a text-processing utility designed to refine Chinese AI-generated text. Analysis of runtime.py and SKILL.md shows the code strictly performs file I/O within the local workspace and utilizes an internal library (skill_runtime.writing_core) for text transformation. There are no indicators of data exfiltration, unauthorized network access, or malicious prompt injection.
能力评估
Purpose & Capability
Name/description match the implementation: runtime.py reads a Markdown/text input, calls humanize_markdown(mode='surgical'), writes a humanized Markdown file and a JSON report into content-production/drafts. The I/O and behavior are proportional to a content-humanizing tool.
Instruction Scope
SKILL.md and README only instruct running the local skill runtime with a specific input file. The runtime only reads the provided input and writes output/report files under workspace_root/content-production/drafts; there are no instructions to read unrelated system files, environment variables, or transmit data to external endpoints.
Install Mechanism
This is an instruction-only skill with no install spec in the manifest (lowest risk). README mentions pip installing requirements.txt, but no requirements.txt is present in the file list — the skill depends on the environment providing skill_runtime.writing_core. Verify that your runtime environment already supplies that module.
Credentials
The skill declares no required environment variables, credentials, or config paths. The only runtime dependency is a workspace_root path (standard for file-based tools). No secret-exposing env vars or unrelated credentials are requested.
Persistence & Privilege
always is false and the skill does not modify other skills or global agent settings. It writes output files into the workspace (expected for a content tool) but does not request elevated or persistent platform privileges.
如何使用
- 确保已安装 OpenClaw(本地或 Docker 部署)
- 在对话框中输入安装命令:
/install content-system-humanizer-zh - 安装完成后,直接呼叫该 Skill 的名称或使用
/content-system-humanizer-zh触发 - 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.0
Initial release of humanizer-zh, a skill for de-AI-ifying Chinese text with strong content constraints:
- Detects and removes obvious AI-writing traces from Chinese prose.
- Surgically rewrites body paragraphs while preserving facts, structure, and author intent.
- Outputs a humanized article copy and a JSON report summarizing changes.
- Strictly avoids inventing facts, altering core arguments, or changing headings.
元数据
常见问题
Content System Humanizer Zh 是什么?
Remove obvious AI-writing traces from Chinese text in a constrained way. Use when Codex needs to reduce AI smell without changing facts, data, or the article... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 192 次。
如何安装 Content System Humanizer Zh?
在 OpenClaw 或 Claude Code 对话框中运行命令「/install content-system-humanizer-zh」即可一键安装,无需额外配置。
Content System Humanizer Zh 是免费的吗?
是的,Content System Humanizer Zh 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。
Content System Humanizer Zh 支持哪些平台?
Content System Humanizer Zh 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。
谁开发了 Content System Humanizer Zh?
由 Abigale-cyber(@abigale-cyber)开发并维护,当前版本 v1.0.0。
推荐 Skills