sero-humanize

审查 · 59
已收录

Audit or edit Sero Markdown documentation and other product prose to remove AI-writing patterns while preserving technical meaning, product terminology, links, examples, and the repository voice. Use when the user asks to humanize, de-slop, tighten, simplify, rewrite, or audit Se

Verified installs0
Stars19
版本1.0.0
质量60/100 · 有潜力
信任59/100 · Do not auto-install
审计74/100 · 需审查

供给资产档案

研究与知识工作

Deep research, source comparison, literature review, RAG, knowledge search, and reports.

浏览赛道

场景

研究 Agent

I need my agent to research a topic, compare sources, and produce a concise report.

适配 Agent

Claude Code + CLI + Codex

适用于 Codex、Claude Code、Cursor、CLI 或自定义 Agent。

安装

就绪

npx skills add sero-labs/sero --skill sero-humanize

维护状态

新鲜

距上次推送 3 天

风险

需审查

Permission surface may require sandboxing

GitHub 质量

19

60/100 质量 · 67/100 信任

覆盖标签

研究研究 Agent安全agent-skill

审查说明

Permission surface may require sandboxing · SKILL.md excerpt is truncated in the provided documentation, but the visible portion is thorough and well-structured.

Agent 采用评分卡

一眼查看信任、审计与安装准备度

这些分数综合公开仓库元数据、OpenAgentSkill 审查信号、维护新鲜度与安装准备度。它用于候选筛选,不替代人工审查。

质量

有潜力
60

有用的候选项,但采用前应与替代方案比较。

信任

Do not auto-install
59

Trust Score v5 found insufficient evidence for agent installation. Treat this as discovery material, not an executable recommendation.

审计

需审查
74

对安装准备度、安全元数据、维护情况与采用风险的机器可读审查。

OpenAgentSkill 信任评分 v5

安装前需人工审查

Choose a stronger alternative or inspect the source manually before any install attempt.

CodexClaude CodeCursorOpenAgentSkill CLI

Stars

19 个 GitHub Stars

仓库活跃度

19 个 Star,1 个 Fork

维护状态

距上次推送 3 天

许可证

Apache-2.0

安装

npx skills add sero-labs/sero --skill sero-humanize

安装安全性

标准软件包或运行时安装路径

权限范围

secrets or environment access, shell or command execution

Agent 结果

暂未有 Agent 结果数据

文档

README/SKILL.md 上下文充分

风险摘要

生产前审查

  • SKILL.md excerpt is truncated in the provided documentation, but the visible portion is thorough and well-structured.
  • Low GitHub adoption signal
  • Quality score needs review
  • Permission surface needs review: secrets or environment access, shell or command execution

安装准备度

安装路径可用

  • 安装路径可用
  • 仓库证据可用
  • 已声明许可证
  • 暂无 Agent 验证结果证据

Agent 可读元数据

这个 Skill 的机器可读决策数据。

使用此区块或内嵌 JSON 判断 Agent 是否应安装该 Skill、选择替代方案,或先请求人工审查。

打开 JSON

适用任务

  • GitHub automation 工作流
  • Claude Code 团队
  • builders willing to evaluate younger projects
  • Inspect repository metadata

适用 Agent

CodexClaude CodeCursorOpenAgentSkill CLICLI

安装决策

命令
npx skills add sero-labs/sero --skill sero-humanize
策略
阻止
人工审查

信任与风险

信任
59/100
审计
74/100
风险级别
需审查

结果闭环

端点
/api/agent/outcome
事件 ID
resolve
结果
5

安装命令

npx skills add sero-labs/sero --skill sero-humanize

不适用场景

  • 需要厂商支持 SLA 的团队
  • production agents without a repository review
  • Low GitHub adoption signal
  • SKILL.md excerpt is truncated in the provided documentation, but the visible portion is thorough and well-structured.
  • 高风险权限提示:Shell or command execution, Secrets or environment access

Agent 安全 v2

34/100 · 避免自动安装

Blocked for auto-install阻止

This skill should not be selected by an agent without explicit human security review.

Do not auto-install. Inspect the source, dependencies, and permission surface first.

通过 API 解析

Shell 或命令执行

Skill 元数据引用了终端、CLI、Shell、子进程或命令执行工作流。

网络访问

Skill 可能访问远程页面、API、仓库或外部服务。

文件系统访问

Skill 可能读取或写入项目文件、文档、生成产物或本地工作区状态。

Secrets or environment access

Skill metadata references credentials, tokens, environment variables, or secret-bearing workflows.

  • 高风险权限提示:Shell or command execution, Secrets or environment access
  • Permission surface may require sandboxing

安装目标

在你的 Agent 工作流中安装此 Skill

通过公开安装端点获取命令、安全清单、目标提示词和该 Skill 的规范链接。

skill install

OpenAgentSkill CLI

Resolve policy, run the source installer safely, and report a verified install receipt.

$ npx --yes https://github.com/Leon-Drq/openagentskill/releases/download/cli-v0.2.1/openagentskill-0.2.1.tgz install sero-labs-sero-humanize

Agent 解析计划

让 Agent 在安装前验证匹配度。

Resolve API 返回首选 Skill、替代方案、安全策略、审计说明、安装目标和可直接执行的提示词,无需抓取此页面。

打开文本计划

Agent 应检查

  • 从 Resolve API 检查任务匹配与替代方案。
  • 检查审计评分、信任评分和安全策略警告。
  • 检查 Codex、Claude Code、Cursor 或 CLI 的安装目标兼容性。

复制提示词

Task: Use sero-humanize in this workspace.
Resolve first: https://www.openagentskill.com/api/agent/resolve?task=Use%20sero-humanize%20for%20an%20agent%20workflow&agent=codex&max_risk=medium
Review install handoff: https://www.openagentskill.com/api/skills/sero-labs-sero-humanize/install
Install command: npx skills add sero-labs/sero --skill sero-humanize
Before running it, summarize audit warnings, required permissions, and the fallback skill if install is risky.

Agent 交接

把安装路径交给 Agent,而不是再给一个目录页。

通过公开安装端点获取命令、安全清单、目标提示词和该 Skill 的规范链接。

打开安装 API

Agent 提示词

Use sero-humanize for this task. Review https://www.openagentskill.com/api/skills/sero-labs-sero-humanize/install, then install with: npx skills add sero-labs/sero --skill sero-humanize

Registry 元数据

用于自动选择 Skill 的 Agent 可读档案。

本页通过 Registry API 提供相同的决策、信任、审计、场景和安装信号,让 Agent 无需抓取界面即可排序。

打开 Manifest

适配 Agent

61/100

GitHub automation

平台

Claude Code

审计报告

需审查 · 74/100

对安装准备度、安全元数据、维护情况与采用风险的机器可读审查。

查看审计报告查看评估报告

Agent 决策面板

Fallback candidate for GitHub automation

先用此 Skill 做原型验证,并保留备选方案。

61
就绪度
原型验证
阶段

栈中角色

备选候选

主要匹配

GitHub automation

信任标签

先做原型验证

安装路径

命令已就绪

适用场景

  • GitHub automation 工作流
  • Claude Code 团队
  • builders willing to evaluate younger projects

证据

  • 仓库近期活跃
  • 已提供安装命令或 GitHub 仓库
  • 60/100 质量档案
  • 6 个 OpenAgentSkill 交互事件

先审查

  • Low GitHub adoption signal
  • SKILL.md excerpt is truncated in the provided documentation, but the visible portion is thorough and well-structured.

实施路径

  1. 1在沙盒 Agent 中安装它,并端到端完成一次GitHub automation任务。
  2. 2Compare output quality, latency, and failure behavior against at least one alternative.
  3. 3Promote it into production only after reviewing repository permissions, license, and maintenance signals.

信任档案

Do not auto-install

Trust Score v5 found insufficient evidence for agent installation. Treat this as discovery material, not an executable recommendation.

59
OpenAgentSkill 信任评分

GitHub 采用度

修复

19 个 GitHub Stars

Star/Fork 活跃度

修复

19 个 Star,1 个 Fork; 当前元数据中没有议题活跃度信息

近期维护

通过

距上次推送 3 天

许可证清晰度

通过

Apache-2.0

积极信号

  • AI 审查已通过
  • 安装路径可用
  • 仓库证据可用
  • 近期维护的仓库
  • 安装命令未发现明显高风险模式
  • 结果闭环已就绪,但需要首次真实 Agent 运行

安装前审查

  • SKILL.md excerpt is truncated in the provided documentation, but the visible portion is thorough and well-structured.
  • Low GitHub adoption signal
  • Quality score needs review
  • Permission surface needs review: secrets or environment access, shell or command execution
  • GitHub adoption: 19 GitHub stars
  • Stars/forks activity: 19 stars, 1 forks; issue activity unavailable in current metadata
  • Permission surface: secrets or environment access, shell or command execution
  • 暂未有真实 Agent 结果报告
  • 无人值守安装前需要人工审查

建议操作

Choose a stronger alternative or inspect the source manually before any install attempt.

质量档案

有潜力 适用于 Agent 工作流的候选

有用的候选项,但采用前应与替代方案比较。

60
GitHub Stars
19
新鲜度
3 天前
安装就绪
许可证
Apache-2.0
安装前审查: Low GitHub adoption signal · SKILL.md excerpt is truncated in the provided documentation, but the visible portion is thorough and well-structured.

工作流匹配

在这些场景使用此 Skill

工作流匹配

加入完整工作流

替代方案短名单

安装前对比

可能适合该任务的相近 Skill。

对比全部

概览

--- name: sero-humanize description: | Audit or edit Sero Markdown documentation and other product prose to remove AI-writing patterns while preserving technical meaning, product terminology, links, examples, and the repository voice. Use when the user asks to humanize, de-slop, tighten, simplify, rewrite, or audit Sero documentation, README text, UI copy, release notes, plans, or specifications for AI tells. Also use when prose needs ASD-STE100 Simplified Technical English. Do not use for code review or for creative and promotional writing. ---

# Sero Humanize

Make Sero prose direct, specific, and useful. Human writing in technical documentation does not need personality. It needs clear decisions, concrete facts, and respect for the reader's time.

## Follow the requested mode

- For an audit or review, report the material patterns and do not edit files. - For an edit, rewrite the named files in place. - For a new document, apply these rules while drafting it. - If the request does not specify a mode, infer it from the requested action. Do not turn a request to assess prose into permission to change it.

## Establish the voice and preservation set

Before editing:

1. Read every file in scope in full. 2. Read the nearest repository instructions that apply to those files. 3. Use adjacent, clearly human-edited Sero documentation as the voice sample when the named files do not establish a consistent voice. 4. Record what must not change: - technical meaning and product behaviour; - product names, canonical terms, and exact UI labels; - commands, code, file paths, numbers, limits, and factual claims; - the documentation type, useful narrative flow, and balance between prose, lists, tables, examples, and callouts; - frontmatter, anchors, link targets, image paths, and screenshot order; - the purpose and placement of each image, diagram, and other media asset; - quotations and user input examples, unless the user asks to edit them.

Identify the intended reader. Unless the page states otherwise, assume the reader knows neither Sero nor the feature. Do not assume that simpler grammar fixes an explanation that requires missing product knowledge.

Do not add a fact to make a sentence more vivid. Verify a doubtful claim from the repository or leave it unchanged and report the doubt.

## Audit structure before wording

Look for clusters and repeated patterns. Do not treat one punctuation mark or one common word as proof of AI writing.

Prioritize these defects:

- Meta narration that announces the next explanation instead of giving it. - Repeated tutorial staging such as "what you are about to learn" and "what you have learned." - Fixed enumerations such as "three things are worth noticing" when a direct heading or short list is clearer. - Several paragraphs that can change order without changing the argument. - A heading followed by a sentence that only repeats the heading. - A conclusion that repeats the introduction without adding an action or fact. - Repeated summaries of the same screen, process, or result. - Forced contrasts such as "not only X, but Y" or "not X; rather Y." - Groups of three used for rhythm instead of meaning. - Mechanical bold lead-ins, excessive inline bold, or lists that should be short prose. - Promotional adjectives, vague importance claims, and unsupported praise. - Vague actors, passive constructions, filler, stacked hedges, and abstract nouns where an action is available. - Synonym cycling for one product concept. Repeat the canonical term. - Long sentences that mix instructions, exceptions, and background. - Em dashes used repeatedly to join thoughts that need separate sentences. - Headings that narrate the demo or expose implementation language instead of naming the reader's task, such as "The finish" or "Answer the gate." - Examples that depend on an unexplained demo domain and therefore do not help the reader understand the feature. - Tutorials that start using the product before they give prerequisites, sample data, expected starting state, or required sign-ins. - Result sections that recite one captured run instead of telling the reader what to inspect and verify.

Keep useful structure. A list, summary, warning, question heading, or em dash is not a defect by itself.

Do not normalize a page or site to one format. Humanize the defective passages, not every paragraph. If a page already explains a concept well in prose, keep it as prose.

## Rewrite for Sero documentation

Apply ASD-STE100 Simplified Technical English where it fits the material:

- Put the action or answer first. - Use active voice when the actor matters. - Give one main instruction per sentence. - Put a condition before the action when the reader must know it first. - Prefer common, precise words over formal or promotional alternatives. - Use the same term for the same thing. - Keep paragraphs focused on one subject. - Keep necessary limits, cautions, and exceptions close to the action. - Use contractions only when the established local voice requires them. - Keep exact UI text in bold when the documentation uses bold for controls. - Keep code identifiers and paths in code formatting. - Retain a summary only when it helps the reader decide or act.

Use lists only when the content is naturally a sequence, set of choices, checklist, or compact reference. Do not:

- convert explanatory prose into bullet points only to make it shorter; - turn each sentence or paragraph into a list item; - replace transitions and reasoning with disconnected bullets; - use repeated lists where a short paragraph gives the reader necessary context; or - make several pages share the same mechanical list structure.

After the sentence pass, read the page as a whole. If lists now dominate a page that previously used useful prose, restore the prose. Clear technical writing needs connected explanation as well as scannable reference material.

For an overview page:

- Explain the feature in familiar words before using its product terms. - State what the user gives Sero, what Sero does, and what the user reviews. - When comparing features, give one plain decision rule. Use examples that a reader can understand without knowing the tutorial repository or a specialist software domain.

For a tutorial:

- Put setup before the first product action. Include required software, accounts, sign-ins, repository or sample-data setup, and a command or visible result that confirms the expected starting state. - Prefer a stable sample repository over instructions that ask an agent to generate approximate sample data. Verify the repository contents and commands before documenting them. - Use task-based headings such as "Review the plan," "Change the plan," and "Check the result." A heading must describe the full purpose of its section; do not narrow a general control to one example case. - End with checks the reader can perform. Do not use a captured run's cost, duration, names, or outcome as a substitute for verification instructions.

For feature language:

- Use the visible object name: icon, button, tab, question, or approval request. Do not call an icon a mark or expose internal terms such as gate, fan-out, or feedback route when plain behaviour is enough. - Keep exact UI labels unchanged, but explain them with common words. - Put high-value quality-of-life features where readers will find them. Give them enough space to explain when the control appears, how to use it, what it changes, and what remains under user control.

Compress or merge only the passages that contain a verified structural defect. Keep useful depth, examples, transitions, and paragraph structure. A shorter page is not automatically a better page.

Do not manufacture a human voice with:

- anecdotes, opinions, jokes, sensory details, or personal asides; - fragments, one-word sentences, or dramatic punch lines; - arbitrary sentence-length variation; - unusual synonyms chosen only to make wording less predictable; - metaphors that replace a precise technical explanation; - deliberate imperfections or tangents; - an invented AI probability or numerical slop score.

## Preserve Markdown and product accuracy

- Do not change fenced code, commands, URLs, link targets, image targets, or frontmatter unless the request requires it. - Do not rename a heading if another page links to its generated anchor without updating that link. - Do not change a UI label to improve prose. Rewrite the surrounding sentence. - Do not remove repetition that is required for independent reference sections. - Do not convert a walkthrough into reference documentation, or reference documentation into a narrative tutorial, without user approval. - Do not infer product behaviour from the prose alone when the edit changes a technical claim. Check the implementation or an authoritative reference. - Treat contradictions between prose, screenshots, capture metadata, sample repositories, and implementation as accuracy defects. Resolve them from the authoritative source instead of rewriting around them. - When a page title changes, update the sidebar, index, related-page labels, and in-scope links that display the old title.

## Preserve images and other media

Treat every existing image, diagram, video, and asset as preserved content. Humanizing prose does not authorize media removal or replacement.

- Do not delete an asset, remove its reference, change its order, or replace it unless the user explicitly approves that action. - Do not use "task value," brevity, a stale appearance, or a text explanation as automatic reasons to remove an image. - Do not bulk-delete assets during a prose revision. - If an image is stale, private, inaccurate, decorative, or duplicated, report the issue and propose one action: keep, recapture, move, or remove. Wait for approval before changing it. - If an image exposes a credential or other active secret, stop publication and report it immediately. Do not silently make a wider set of image changes. - When a replacement is approved, capture or obtain the replacement before removing the current asset. Preserve the route and layout while replacement work is pending. - Check non-doc consumers before changing an asset. README files, homepages, package pages, and other applications can import docs-site images directly.

A decision not to add a new screenshot is not permission to remove an existing screenshot.

## Control the size of the rewrite

For a large documentation set, work in reviewed vertical slices. Complete and review one representative page before applying the approach to the rest of a slice. Do not perform a site-wide structural rewrite from an audit summary.

Pause and ask for approval when the work would:

- change the dominant format of a page, such as prose to lists; - remove substantial explanation, examples, or media; - merge, tombstone, redirect, or delete a page; - change many pages through the same structural template; or - produce a much larger diff than the factual and prose defects require.

When several agents contribute, give them the same preservation set and require a central review of format balance and media changes before integration.

## Use a two-pass edit

### Pass 1: structure

Remove redundant framing, merge repeated explanations, order information by the reader's task, and keep prerequisites before dependent actions. Give prominent placement to features that materially improve repeated use; do not give every feature equal weight merely because the source page did.

Keep the smallest effective structural change. Do not rewrite a complete page when a heading, transition, or paragraph edit fixes the defect.

### Pass 2: sentences

Remove filler and AI mannerisms. Simplify grammar. Keep terminology and facts stable. Read the result as technical documentation, not as marketing copy.

Then compare the result with th

技术详情

版本
1.0.0
许可证
Apache-2.0
最近更新
2026年8月19日
发布时间
2026年8月19日

决策摘要

备选候选

61
就绪
原型验证
阶段

仓库近期活跃

审计

安装审查

安装与采用审查

74
需审查
安全性
74/100
维护状态
100/100
安装
92/100
打开完整审计查看评估报告

Agent 验证证据

Agent 验证证据

来自解析、审查、安装和一次小范围运行后的结果报告。

0
已验证
Needs first agent run自动安装: 先审查最近: 未知
成功率
近期失败
结果
0
输出质量
失败
0
不相关
0
安装次数
0
风险拦截
0
需要配置
0
生产环境
0

暂时没有 Agent 结果数据。首次 Agent 执行可以通过 /api/agent/outcome 报告成功、需要设置、风险拦截、失败或不相关。

安装

加入 Agent 工作流

免费且开源. 在生产 Agent 中安装前请先审查报告。

增长闭环

分享工具包

X

为 sero-humanize 准备的场景化草稿,可手动发布到 X。

策展说明
A practical pick for design or creative work:

sero-humanize: Audit or edit Sero Markdown documentation and other product prose to remove AI-writing patterns while preserving technical...

19 stars

https://www.openagentskill.com/skills/sero-labs-sero-humanize?ref=x
打开 X 草稿
可选:带安装命令的回复
Listing + install path for sero-humanize:
https://www.openagentskill.com/skills/sero-labs-sero-humanize?ref=x

Install: npx skills add sero-labs/sero --skill sero-humanize
打开回复草稿

收录来源

Registry 收录

可认领

此列表来自公开来源,维护者认领获批前不会标记为官方。

创作者
sero-labs
收录方
OpenAgentSkill 社区索引

归属链接指向公开仓库或创作者主页。创作者可认领列表以更新所有权信号。

认领此 Skill

所有者认领

认领此 Skill 页面

这条 Registry 收录 列表归属于 sero-labs,但尚未标记为官方。认领后可增加已验证所有者信号,使后续发布、安装和审计更新更值得信赖。

创作者外链工具包

将证据徽章加入你的 README

在开发者评估仓库的位置展示规范页面、当前信任与审计信号,以及真实的 Agent 验证证据。

[![Listed on OpenAgentSkill](https://www.openagentskill.com/api/badge/sero-labs-sero-humanize?metric=listed&label=Listed)](https://www.openagentskill.com/skills/sero-labs-sero-humanize)
[![OpenAgentSkill Trust](https://www.openagentskill.com/api/badge/sero-labs-sero-humanize?metric=trust&label=Trust)](https://www.openagentskill.com/skills/sero-labs-sero-humanize)
[![OpenAgentSkill Audit](https://www.openagentskill.com/api/badge/sero-labs-sero-humanize?metric=audit&label=Audit)](https://www.openagentskill.com/skills/sero-labs-sero-humanize/audit)
[![Agent Proven](https://www.openagentskill.com/api/badge/sero-labs-sero-humanize?metric=proven&label=Agent%20Proven)](https://www.openagentskill.com/skills/sero-labs-sero-humanize)

作者

S

sero-labs

@sero-labs

平台适配

健康信号

GitHub Stars
19
质量评分
33/100
最近 GitHub 推送
2026年8月19日
框架提示
未知
OpenAgentSkill 浏览量
6
复制安装命令
0
跳转点击
0

社区信号

告诉我们这个 Skill 是否对你的 Agent 工作流有帮助。汇总反馈会持续改善排序。

信任与安全

Do not auto-install

59
  • GitHub 采用度19 个 GitHub Stars修复
  • Star/Fork 活跃度19 个 Star,1 个 Fork; 当前元数据中没有议题活跃度信息修复
  • 近期维护距上次推送 3 天通过
  • 许可证清晰度Apache-2.0通过
  • README/SKILL.md 完整度元数据包含足够的用法与工作流上下文通过
  • 依赖与运行时风险凭据或环境变量访问信息