novel-standard

审查 · 70
已收录

Standard writing workflow in which the main Agent drafts and revises while the built-in general-purpose SubAgent reviews, balancing quality and speed.

Verified installs0
Stars647
版本1.0.0
质量75/100 ·
信任70/100 · 仅限沙盒
审计83/100 · 可安全尝试

供给资产档案

编程与开发 Agent

代码审查、仓库分析、测试、CI、GitHub、DevOps 与开发工作流 Skill。

浏览赛道

场景

GitHub automation

I need my agent to triage GitHub issues, review pull requests, and summarize repository changes.

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 alfredxw/denova --skill novel-standard

维护状态

新鲜

今天有推送

风险

可安全尝试

The provided SKILL.md excerpt is truncated, but the visible content is complete and coherent.

GitHub 质量

647

75/100 质量 · 78/100 信任

覆盖标签

编程GitHub automationwritingagent-skill

审查说明

The provided SKILL.md excerpt is truncated, but the visible content is complete and coherent. · Quality score needs review

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

75

可靠的选择,值得加入生产工作流候选列表。

信任

仅限沙盒
70

有用但信任信号不足或混杂的候选项。在结果闭环证明任务匹配前,请保持在隔离工作区内使用。

审计

可安全尝试
83

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

仅在沙盒中运行,并在用于真实工作前比较接近的替代方案。

CodexClaude CodeCursorOpenAgentSkill CLI

Stars

647 个 GitHub Stars

仓库活跃度

647 个 Star,111 个 Fork

维护状态

今天有推送

许可证

Apache-2.0

安装

npx skills add alfredxw/denova --skill novel-standard

安装安全性

标准软件包或运行时安装路径

权限范围

文件系统或文档访问

Agent 结果

暂未有 Agent 结果数据

文档

Usable metadata, review docs

风险摘要

生产前审查

  • The provided SKILL.md excerpt is truncated, but the visible content is complete and coherent.
  • Quality score needs review

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

  • Local desktop 工作流
  • Claude Code 团队
  • 重视 GitHub 采用信号的团队
  • Navigate local resources

适用 Agent

CodexClaude CodeCursorOpenAgentSkill CLICLI

安装决策

命令
npx skills add alfredxw/denova --skill novel-standard
策略
审查
人工审查

信任与风险

信任
70/100
审计
83/100
风险级别
可安全尝试

结果闭环

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

安装命令

npx skills add alfredxw/denova --skill novel-standard

不适用场景

  • 需要厂商支持 SLA 的团队
  • production agents without a repository review
  • The provided SKILL.md excerpt is truncated, but the visible content is complete and coherent.
  • 暂未有 OpenAgentSkill 使用反馈数据
  • Quality score needs review

Agent 安全 v2

63/100 · 安装前审查

已审查并附权限说明审查

可用候选,但 Agent 在安装前应展示权限与审计说明。

在真实工作区安装前需要人工批准。

通过 API 解析

Browser automation

Skill may drive a browser or interact with web pages.

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。

  • The provided SKILL.md excerpt is truncated, but the visible content is complete and coherent.

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 alfredxw-novel-standard

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 novel-standard in this workspace.
Resolve first: https://www.openagentskill.com/api/agent/resolve?task=Use%20novel-standard%20for%20an%20agent%20workflow&agent=codex&max_risk=medium
Review install handoff: https://www.openagentskill.com/api/skills/alfredxw-novel-standard/install
Install command: npx skills add alfredxw/denova --skill novel-standard
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 novel-standard for this task. Review https://www.openagentskill.com/api/skills/alfredxw-novel-standard/install, then install with: npx skills add alfredxw/denova --skill novel-standard

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

86/100

Local desktop

平台

Claude Code

审计报告

可安全尝试 · 83/100

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

适合 Local desktop 的首选

将其作为优先候选,再在你的 Agent 环境中验证 README 与安装路径。

86
就绪度
采用
阶段

栈中角色

首选

主要匹配

Local desktop

信任标签

可用于生产

安装路径

命令已就绪

适用场景

  • Local desktop 工作流
  • Claude Code 团队
  • 重视 GitHub 采用信号的团队

证据

  • 647 个 GitHub Stars
  • 仓库近期活跃
  • 已提供安装命令或 GitHub 仓库
  • 75/100 质量档案

先审查

  • The provided SKILL.md excerpt is truncated, but the visible content is complete and coherent.
  • 暂未有 OpenAgentSkill 使用反馈数据

实施路径

  1. 1在沙盒 Agent 中安装它,并端到端完成一次Local desktop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

仅限沙盒

有用但信任信号不足或混杂的候选项。在结果闭环证明任务匹配前,请保持在隔离工作区内使用。

70
OpenAgentSkill 信任评分

GitHub 采用度

信息

647 个 GitHub Stars

Star/Fork 活跃度

信息

647 个 Star,111 个 Fork; 当前元数据中没有议题活跃度信息

近期维护

通过

今天有推送

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
  • 有意义的 GitHub 采用信号
  • 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

  • The provided SKILL.md excerpt is truncated, but the visible content is complete and coherent.
  • Quality score needs review
  • 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

仅在沙盒中运行,并在用于真实工作前比较接近的替代方案。

质量档案

适用于 Agent 工作流的候选

可靠的选择,值得加入生产工作流候选列表。

75
GitHub Stars
647
新鲜度
今天
安装就绪
许可证
Apache-2.0
安装前审查: The provided SKILL.md excerpt is truncated, but the visible content is complete and coherent.

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: novel-standard description: Standard writing workflow in which the main Agent drafts and revises while the built-in general-purpose SubAgent reviews, balancing quality and speed. category: writing capabilities: - writing-workflow agent: ide ---

# novel-standard

Use this standard IDE Writing Agent workflow to balance quality and speed.

## Determine the writing scope

- Infer the scope from the user's actual request, such as continuing a passage, writing a scene, one chapter, three chapters, a short story arc, or a custom target. - Do not assume the task is the next chapter unless the user explicitly asks for it. - There is no `writing_scope` field. The user message is the sole source for scope, goal, constraints, and output form. - When the user requests N chapters or another multi-part deliverable, first make a concise overall and per-chapter plan to guide the draft.

## Workflow

main Agent drafts -> general-purpose SubAgent reviews -> main Agent revises and updates state -> final output

Use only the main Agent and the built-in `general-purpose` SubAgent. Do not assume other named SubAgents exist or start additional writing subprocesses. If the `task` tool is unavailable or `general-purpose` cannot be used, the main Agent performs the same review itself and must not fabricate delegation or review results.

## Tool requirements

- Before writing, use `read` for the necessary context: `CREATOR.md`, `setting/outline.md`, `setting/progress.md`, `setting/character-states.md`, the relevant chapter-group plan, and recent chapters. For lore, use `list_lore_items` to identify relevant entries and `read_lore_items` to load their complete bodies. - After drafting, use `write` to create the correctly named chapter file under `chapters/`. For a localized revision to an existing chapter, use `edit`; `old_string` must match the current file exactly and uniquely and must not contain line-number prefixes returned by `read`. - When available, delegate review to `general-purpose` through `task`. Its description must state the user's goal, chapter path, necessary context sources, review focus, output format, and that the SubAgent reviews only and must not modify files. - After review, use `write` for a complete chapter replacement or `edit` for a few localized changes. Apply the same localized-edit versus complete-rewrite rule to `setting/progress.md` and `setting/character-states.md`. - Inspect every `write` or `edit` result. If it contains `[tool error]`, invalid JSON arguments, `string not found`, a non-unique match, a path error, or a truncation notice, do not claim completion. Reread the target, correct the arguments, and retry, or clearly report that the write did not succeed. - Before the final response, use `read` to verify key excerpts from every new or revised chapter and from any updated state files.

1. Draft at the scope and under the constraints requested by the user, then write the draft to the correctly named chapter file under `chapters/`. Do not update progress or character state yet. 2. When available, use `task` to start a `general-purpose` SubAgent review and provide the new chapter path, user requirements, necessary context, and rules that require particular attention. Otherwise, perform the review in the main Agent. 3. The review returns structured issues only and does not edit prose. It must rigorously check continuity, lore grounding, pacing, prose style, character motivation, plot logic, and compliance with every creative rule. Do not include praise. 4. Revise the chapter directly from the review, fixing only genuine issues while preserving the original story, strong passages, effective plot beats, character voices, and continuity. 5. After the final revision, update `setting/progress.md` and `setting/character-states.md` in the same turn without waiting for separate chapter confirmation. Chapter status is only a UI editing marker. Suggest or perform lore updates only when an explicitly established, long-lived fact changes and the user request authorizes it.

## Review requirements

The `general-purpose` review or main-Agent self-review may read the necessary preceding prose, `CREATOR.md`, outline, progress, character state, and lore for comparison. Check the new chapter against the task, user prompt, `CREATOR.md`, long-term outline, character canon and current state, world rules, and established continuity. Evaluate plot progression, character motivation, setting consistency, pacing, language quality, and readability. Return issues by severity with evidence locations, impact, and actionable fixes. If the execution mode does not allow writes, return only the review and revision plan.

## Final output

- Return the final prose or other writing artifact requested by the user. - Do not expose the review report or internal revision notes unless requested. - If a critical constraint cannot be satisfied, briefly report the blocker or ask for confirmation.

技术详情

版本
1.0.0
许可证
Apache-2.0
最近更新
2026年8月22日
发布时间
2026年8月22日

决策摘要

首选

86
就绪
采用
阶段

647 个 GitHub Stars

审计

安装审查

安装与采用审查

83
可安全尝试
安全性
83/100
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 novel-standard 准备的场景化草稿,可手动发布到 X。

策展说明
A practical pick for a repeatable workflow:

novel-standard: Standard writing workflow in which the main Agent drafts and revises while the built-in general-purpose SubAgent reviews, b...

647 stars

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

Install: npx skills add alfredxw/denova --skill novel-standard
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
alfredxw
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 alfredxw,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/alfredxw-novel-standard?metric=listed&label=Listed)](https://www.openagentskill.com/skills/alfredxw-novel-standard)
[![OpenAgentSkill Trust](https://www.openagentskill.com/api/badge/alfredxw-novel-standard?metric=trust&label=Trust)](https://www.openagentskill.com/skills/alfredxw-novel-standard)
[![OpenAgentSkill Audit](https://www.openagentskill.com/api/badge/alfredxw-novel-standard?metric=audit&label=Audit)](https://www.openagentskill.com/skills/alfredxw-novel-standard/audit)
[![Agent Proven](https://www.openagentskill.com/api/badge/alfredxw-novel-standard?metric=proven&label=Agent%20Proven)](https://www.openagentskill.com/skills/alfredxw-novel-standard)

作者

A

alfredxw

@alfredxw

平台适配

健康信号

GitHub Stars
647
质量评分
43/100
最近 GitHub 推送
2026年8月22日
框架提示
未知
OpenAgentSkill 浏览量
0
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

仅限沙盒

70
  • GitHub 采用度647 个 GitHub Stars信息
  • Star/Fork 活跃度647 个 Star,111 个 Fork; 当前元数据中没有议题活跃度信息信息
  • 近期维护今天有推送通过
  • 许可证清晰度Apache-2.0通过
  • README/SKILL.md 完整度公开元数据需要更完整的 README/SKILL.md 上下文信息
  • 依赖与运行时风险公开元数据中未发现主要依赖风险提示通过