Agent skill

Story Setup

by Xiaoyangy in Xiaoyangy/novel-studio

网文写作工具集基础设施部署。将 hooks/rules/agents/CLAUDE.md/AGENTS.md 等基础设施部署到用户项目目录,支持 Claude Code / OpenCode / Codex / OpenClaw。触发方式:/story-setup、$story-setup、「准备写书」「帮我搭一下环境」「配置写作项目」。

Apache-2.0Auto-check passedAgent Workflows

Install Story Setup

skills CLI
$ npx skills add Xiaoyangy/novel-studio --skill story-setup -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install Xiaoyangy/novel-studio story-setup --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/Xiaoyangy/novel-studio.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/story-setup .claude/skills/story-setup && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
story-setup
GitHub stars
131
Token cost
~5.7k tokens
SKILL.md length
1,744 words
Files
92 (incl. references)
Skills in repo
18
Repo updated
First seen
Licence
Apache-2.0

At a glance

网文写作工具集基础设施部署。将 hooks/rules/agents/CLAUDE.md/AGENTS.md 等基础设施部署到用户项目目录,支持 Claude Code / OpenCode / Codex / OpenClaw。触发方式:/story-setup、$story-setup、「准备写书」「帮我搭一下环境」「配置写作项目」。

  • Works in 3 steps: :检测项目状态 → :部署基础设施 → :验证安装
  • Tasks that involve Agent instruction files
  • SKILL.md covers Phase 1:检测项目状态, Phase 2:部署基础设施, Codex hooks.json… and OpenClaw skills-only…, plus 8 more sections
  • Calls opencode

What it does

Story Setup is an agent skill from Xiaoyangy/novel-studio. 网文写作工具集基础设施部署。将 hooks/rules/agents/CLAUDE.md/AGENTS.md 等基础设施部署到用户项目目录,支持 Claude Code / OpenCode / Codex / OpenClaw。触发方式:/story-setup、$story-setup、「准备写书」「帮我搭一下环境」「配置写作项目」。

Its SKILL.md is about 5.7k tokens, which your agent loads only when the skill is triggered. The skill folder holds 93 other files, including reference files (for example `CONTEXT.md`, `UPGRADING.md` and `context.json`).

It sits in Agent Workflows, covering Agent instruction files and Creative writing and fiction. The repository describes itself as: 开源、本地优先的 AI 长篇小说创作引擎:多智能体世界推演、按弧规划、RAG 长程记忆、逐章审核与断点恢复 | Open-source AI novel generator for long-form fiction. The licence is Apache-2.0.

When your agent uses it

  • Tasks that involve Agent instruction files
  • Tasks that involve Creative writing and fiction

Example prompts

  • “/story-setup”

Requirements

  • Python 3

Workflow steps

3 steps, taken from the step headings in SKILL.md.

  1. :检测项目状态
  2. :部署基础设施
  3. :验证安装

What it can do on your machine

Read from SKILL.md and the folder at commit ed04a10. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • opencode

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    Links to these hosts (documentation or services it may open):

    • models.dev

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Story Setup loads about 5.7k tokens when it runs, and up to ~201k if it reads all its reference files. Until then it costs about 46 tokens; SKILL.md has 1,744 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~46
When it runs · the whole SKILL.md, loaded when a task matches
~5.7k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~201k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from Xiaoyangy/novel-studio at commit ed04a10, republished under its Apache-2.0 licence (© Xiaoyangy). 1,744 words, ~5,711 tokens.

Download SKILL.mdSave it as .claude/skills/story-setup/SKILL.md (or your agent's skills folder). This skill also uses 91 other files; get the full folder from GitHub.
name
story-setup
description
网文写作工具集基础设施部署。将 hooks/rules/agents/CLAUDE.md/AGENTS.md 等基础设施部署到用户项目目录,支持 Claude Code / OpenCode / Codex / OpenClaw。触发方式:/story-setup、$story-setup、「准备写书」「帮我搭一下环境」「配置写作项目」。
version
1.2.5

story-setup:网文写作工具集基础设施部署

你是写作基础设施部署器。将网文写作工具集的全套基础设施(hooks、rules、agents、CLAUDE.md、AGENTS.md、Codex/OpenClaw 配置)部署到用户项目目录。

执行铁律:不覆盖用户已有配置,合并而非替换。


Phase 1:检测项目状态

  1. 检查当前目录是否已部署过(存在 .story-deployed)
    • 如果已存在 → 使用 AskUserQuestion 确认是否重新部署
  2. 检查是否有书名目录(包含 追踪/ 子目录的目录,或用户自定义结构)
    • 有 → 识别为长篇项目,显示当前项目信息
    • 无 → 识别为新项目或短篇项目
  3. 检查 .claude/settings.local.json 是否存在
    • 存在 → 读取现有配置,后续合并
    • 不存在 → 后续创建新文件
  4. 检查 .active-book 文件是否存在
    • 存在 → 显示当前活跃书目
    • 不存在 → 跳过
  5. 检查 opencode.json 或 .opencode/ 是否存在
    • 存在 → 识别为 opencode 项目,target_cli = opencode
    • 不存在 → 跳过
  6. 检查 .codex/、.codex/config.toml、.codex/agents/、.codex/hooks.json、AGENTS.md 中的 Codex 段
    • 存在 → 识别为 Codex 项目,target_cli = codex
    • 不存在 → 跳过
  7. 检查 openclaw.json、.openclaw/、.agents/skills/、AGENTS.md 中的 OpenClaw 段,或 skills/*/SKILL.md 中的 metadata.openclaw
    • 存在 → 识别为 OpenClaw 项目,target_cli = openclaw
    • 不存在 → 跳过
  8. 如 .claude/ 或 CLAUDE.md、opencode 标记、Codex 标记、OpenClaw 标记同时存在 → 使用 AskUserQuestion 让用户选择目标 CLI(选项:仅 Claude Code / 仅 OpenCode / 仅 Codex / 仅 OpenClaw / 任意组合)
  9. 如四者都不存在(全新项目)→ 使用 AskUserQuestion 让用户选择目标 CLI
    • 用户选择 opencode → target_cli = opencode,部署时创建 opencode.json 和 .opencode/
    • 用户选择 claude-code → 按现有逻辑处理
    • 用户选择 codex → target_cli = codex,部署时创建 .codex/
    • 用户选择 openclaw → target_cli = openclaw,部署时复制 OpenClaw 兼容 skills 到项目 skills/
    • 用户选择多端 → target_cli = claude-code,opencode,codex,openclaw(仅包含用户选择的端)

Phase 2:部署基础设施

使用 AskUserQuestion 确认部署位置后,依次执行。

2.0 部署清单(机械可检查)
Source pathTarget pathOwner classMerge modeValidation check
skills/story-setup/references/templates/CLAUDE.md.tmplCLAUDE.mduser+managedmarker/section mergecontains story skill routing sections
skills/story-setup/references/templates/hooks/.claude/hooks/story-setup managedrecursive replacesession-*.sh, detect-story-gaps.sh, validate-story-commit.sh, guard-outline-before-prose.sh, check-prose-after-write.sh, lib/common.sh, lib/sentinel.sh exist
skills/story-setup/references/templates/rules/*.md.claude/rules/*.mdstory-setup managedreplaceevery rule contains paths frontmatter
skills/story-setup/references/templates/agents/*.md.claude/agents/*.mdstory-setup managedreplace7 agent files exist
skills/story-setup/references/agent-references/*.md.claude/skills/story-setup/references/agent-references/*.mdstory-setup managedreplaceevery story-setup/references/agent-references/*.md reference resolves
skills/story-setup/references/templates/settings-hooks.json.claude/settings.local.jsonuser+managedmerge by hook commandhook JSON valid and registered commands deduped
skills/story-setup/references/templates/上下文.md.tmpl{书名}/追踪/上下文.mduser statecreate only if absentnever overwrite existing writing context
generated sentinel.story-deployedstory-setup managedreplacecontains agents_version, setup_skill_version, target_cli, resolver_strategy, references_dir
skills/story-setup/references/opencode/AGENTS.md.tmplAGENTS.mduser+managedmarker/section mergecontains story skill routing sections
skills/story-setup/references/opencode/agents/.opencode/agents/story-setup managedreplace7 agent files exist(replace 前按 2.4.4 Step 0 缓存现有 model:,避免覆盖用户已配模型)
skills/story-setup/references/opencode/plugin.ts.opencode/plugins/story-hooks.tsstory-setup managedreplaceTypeScript plugin file exists
skills/story-setup/references/opencode/commands/.opencode/commands/story-setup managedreplace13 command files exist
skills/story-setup/references/opencode/opencode.json.patchmerge into opencode.jsonuser+managedmerge by plugin/permission keyplugin entry registered
skills/story-setup/references/agent-references/skills/story-setup/references/agent-references/story-setup managedreplaceevery reference resolves
skills/story-setup/references/opencode/pre-commit.sh.git/hooks/pre-commituser+managedappend or createfile exists and is executable;含 marker 块则替换块内容,不含则检测 exit 0 位置智能插入
skills/story-setup/references/codex/AGENTS.md.tmplAGENTS.mduser+managedmarker/section mergecontains Codex story skill routing sections
skills/story-setup/references/codex/agents/.codex/agents/story-setup managedreplace7 TOML agent files parse and contain name/description/developer_instructions
skills/story-setup/references/codex/hooks/hooks.json.codex/hooks.jsonuser+managedmerge by event+commandhook JSON valid; commands deduped
skills/story-setup/references/codex/hooks/story_codex_hook.py.codex/hooks/story_codex_hook.pystory-setup managedreplacePython syntax valid
skills/story-setup/references/agent-references/.codex/skills/story-setup/references/agent-references/story-setup managedreplaceevery reference resolves
skills/story-setup/references/openclaw/AGENTS.md.tmplAGENTS.mduser+managedmarker/section mergecontains OpenClaw story skill routing sections
repository skills/story*/skills/story*/story-setup managed for known skill namesreplace known skill dirs only8 SKILL.md files exist; OpenClaw-compatible frontmatter
skills/story-setup/references/agent-references/skills/story-setup/references/agent-references/story-setup managedreplace via full skill copyevery reference resolves
opencode.json 合并算法

部署 opencode.json.patch 时按以下规则合并:

  1. 读取现有 opencode.json(如存在),解析 JSON
  2. 合并 plugin 数组:将 ./.opencode/plugins/story-hooks.ts 加入数组,去重
  3. 保留用户已有的其他配置字段(permission、model、provider 等),不覆盖
  4. 写入合并后的 opencode.json
2.1 部署 CLAUDE.md
  • 读取 skills/story-setup/references/templates/CLAUDE.md.tmpl
  • 替换占位符(见下方「模板占位符」段)
  • 写入项目根目录 CLAUDE.md(如已存在,按「CLAUDE.md 合并策略」处理)
2.2 部署 Hooks
  • 递归复制完整目录树:将 skills/story-setup/references/templates/hooks/ 复制到用户项目 .claude/hooks/
  • 必须保留子目录 lib/,其中:
    • lib/common.sh 提供 project_root、discover_active_book、discover_all_books
    • lib/sentinel.sh 提供 .story-deployed 字段读取
  • 只需对 .claude/hooks/*.sh 设置执行权限(chmod +x);lib/*.sh 由 hook source,不要求可执行位
2.3 部署 Rules
  • 读取 skills/story-setup/references/templates/rules/ 下所有 .md 文件
  • 复制到用户项目的 .claude/rules/ 目录
2.4 部署 Agents
  • 读取 skills/story-setup/references/templates/agents/ 下所有 .md 文件
  • 复制到用户项目的 .claude/agents/ 目录
  • Agent 文件属于 story-setup 管理文件,可安全覆盖;版本升级时按 UPGRADING.md 的版本检测结果重新部署
  • 部署后必须新开会话:Claude Code 只在会话启动时扫描 .claude/agents/ 注册 subagent。当前会话内新部署的 agent 不会立即可用——必须让用户新开一个 Claude Code 会话,story-architect/narrative-writer 等 custom agent 才会注册成 subagent_type;否则 story-review、story-long-write 等想 spawn 时会拿到「subagent_type 不可用」并降级 solo(单视角)。这一步必须在安装报告里明确告知用户(见 Phase 3 第 6 步)。
2.4.1 Agent 兼容性处理
  • Agent frontmatter 以 Claude Code 为主;OpenCode 由 scripts/sync-opencode.py 生成 .opencode/agents/*.md;Codex 由 scripts/generate-codex-agents.py 生成 .codex/agents/*.toml。
  • OpenClaw Phase 1 不部署 agents:OpenClaw 只部署 skills,agent 协作相关 skill 必须按既有 fallback 规则降级 solo/direct,不要把 Claude/OpenCode agent frontmatter 直接复制成 OpenClaw agent。
  • 部署到项目后,agent 内引用的参考资料必须走 story-setup/references/agent-references/*.md 这一本 skill 内复制路径;不要跨 skill 引用其他 skill 的 references。若全局安装路径不同,优先用项目内 .claude/skills/ 或 skills/ 作为规范路径前缀,其次用工具的 skill 搜索能力,不要假定固定绝对路径。
2.4.2 部署 Agent References
  • 将 skills/story-setup/references/agent-references/ 下所有 .md 复制到项目内 .claude/skills/story-setup/references/agent-references/
  • 如目标项目已经使用项目本地 skills/ 目录,也可以同步复制到 skills/story-setup/references/agent-references/ 作为 fallback,但不得只复制 fallback 而遗漏 .claude/skills/ 主路径
  • 校验:凡 agent 或 reference 中出现 story-setup/references/agent-references/<file>.md,源包与目标包都必须存在 <file>.md
2.4.3 部署 Codex Agents(target_cli 含 codex 时)
  • 读取 skills/story-setup/references/codex/agents/ 下所有 .toml 文件,复制到用户项目 .codex/agents/
  • Agent 文件属于 story-setup 管理文件,可安全覆盖;生成源由 scripts/generate-codex-agents.py 从 Claude agent 模板确定性生成
  • 校验每个 TOML 都能解析,且包含 Codex 必需字段:name、description、developer_instructions
  • 只读职责 agent(consistency-checker、story-explorer)必须保留 sandbox_mode = "read-only"
  • 部署后必须 trust + 新开 Codex 会话:Codex custom agents 位于 .codex/agents/*.toml,项目 .codex/ 配置层需要被 trust;部署后需要新会话/刷新后才可能稳定暴露给 spawn。若运行时返回 unknown agent_type,调用方必须降级 solo/direct 并报告 fallback。
  • 将 skills/story-setup/references/agent-references/ 同步复制到 .codex/skills/story-setup/references/agent-references/,作为 Codex agent 的项目内参考资料主路径
2.4.4 配置 OpenCode Agent 模型

仅当 target_cli 含 opencode 时执行。OpenCode 子代理不指定模型时继承主模型,导致低成本 Agent 也消耗主模型额度。此步骤自动检测用户模型并写入 model: 字段。

Step 0:保留已有模型配置(必须在 .opencode/agents/ 的 replace 之前执行)

OpenCode agents 部署是 replace,会覆盖上次写入的 model:。所以在执行该 replace 之前先扫描现有 .opencode/agents/*.md,缓存每个 agent 的 model:(agent 名 → 模型 ID)。后续检测失败/超时、或用户跳过某一级时,用缓存值回填,避免把用户上次配好的低成本模型抹成主模型。若 replace 已先发生、缓存为空,则按全新部署处理,并在安装报告中提示"未能保留上次模型配置"。

Step 1:获取模型列表

优先执行 opencode models --verbose,它输出含 cost(input/output/cache 单价)、context、capabilities 的 metadata;不可用或解析失败时回退到 opencode models 纯文本(每行 provider/model)。两者都用 60000ms(60 秒)超时,因为首次运行需加载 models.dev 缓存。

  • 成功 → 进入 Step 2
  • 超时 → 重试一次(缓存可能未预热);仍然超时则按 Step 0 缓存回填已有 model:、跳过自动配置,在安装报告中输出手动配置指南
  • 失败(命令不存在、输出为空等)→ 同上:回填 Step 0 缓存、跳过自动配置、输出手动配置指南
Step 2:模型分级

优先按成本分级(有 --verbose 时):按每模型实际 cost 从低到高分档——低端取最便宜/免费档、中端取中价档、高端取最贵或上下文/能力最强档。免费模型按真实 cost=0 归低端,不按名字里的营销词(如 nemotron-3-ultra-free 名含 ultra 但 cost=0,应归低端)。无 cost 数据的模型也据此进入候选,不被丢弃。

回退按关键词分级(无 --verbose 或无 cost 时):按模型 ID 中最后一个 / 之后的模型名按 -、.、_ 分割为段,逐段精确匹配关键词(不区分大小写)。例如 minimax-m3 拆为 [minimax, m3],不匹配 mini 也不匹配 max;claude-haiku-4.5 拆为 [claude, haiku, 4, 5],匹配 haiku。关键词分级是启发式,安装报告中标注 分级依据:关键词(heuristic)。

等级匹配关键词对应 Agent
低端haiku, flash, mini, nano, liteconsistency-checker, story-explorer
中端sonnet, plusstory-researcher, narrative-writer, character-designer
高端opus, pro, ultra, maxstory-architect
  • 一个模型可能匹配多个等级的关键词,取最高等级
  • 关键词回退下未匹配任何关键词的模型仍列入候选附加建议(按成本分级则一律纳入),并在安装报告列出,提示"可通过自定义输入使用"
  • 同一等级内,如果包含多个模型供应商,优先列出知名供应商(anthropic、openai、google、deepseek)的模型
Step 3:逐级交互选择

按 低端 → 中端 → 高端 顺序,每级用 AskUserQuestion 让用户选择。

低端选项结构:

问题:"为低成本 Agent(consistency-checker, story-explorer)选择模型:"
选项:
  - provider/model-id
  - provider/model-id
  - 自定义输入(手动输入完整模型 ID,ID 拼写错误要到运行时才会暴露)
  - 跳过,使用主模型(成本可能较高)

中端选项结构:

问题:"为写作质量关键 Agent(narrative-writer, character-designer, story-researcher)选择模型:"
选项:
  - provider/model-id
  - provider/model-id
  - 自定义输入(请勿使用低端模型,会影响正文质量;ID 拼写错误要到运行时才会暴露)
  - 跳过,使用主模型(主模型质量通常足够)

高端选项结构:

问题:"为总指挥 Agent(story-architect)选择模型:"
选项:
  - provider/model-id
  - provider/model-id
  - 自定义输入(手动输入完整模型 ID,ID 拼写错误要到运行时才会暴露)
  - 跳过,使用主模型(成本可能较高)

规则:

  • 候选最多显示 5 个,超过则截断并提示"更多模型请使用自定义输入"。每一级无论候选数是否为 0 都用 AskUserQuestion 弹出,选项至少含:候选模型(如有)、自定义输入、保留现有模型(Step 0 缓存到该 agent 的 model,无则不显示此项)、跳过,用主模型。候选为 0 时仍弹窗,并在问题说明里给出对应警告 + 列出未分级/未入档模型供参考——不再静默跳过交互(否则用户够不到自定义输入)。
  • 自定义输入:用户输入 provider/model-id 完整 ID;写入前校验为单行、无控制字符、匹配 ^[A-Za-z0-9._-]+/[A-Za-z0-9._:+-]+$,不符则提示重输或改选跳过。
  • 保留现有模型:写回 Step 0 缓存的该 agent model(重新部署时保住用户上次配置),不算"跳过"。
  • 跳过,用主模型:显式清除——不写该 agent 的 model:,agent 继承主模型。想保留上次配置请选 保留现有模型。
  • 各级候选为 0 时在问题说明里给出提示:
    • 低端:"未检测到低成本模型,这 3 个 agent 将使用主模型,成本可能较高"
    • 中端:"未检测到匹配的中端模型。narrative-writer、character-designer、story-researcher 将使用主模型。如主模型质量足够此配置合理;如需降本,请用自定义输入指定不低于主模型质量的中端模型,或从下方未分级模型里选。"
    • 高端:"未检测到高端模型,story-architect 将使用主模型"
Step 4:写入 model 字段

对应用户选择的 agent 文件(.opencode/agents/*.md,由部署清单中 OpenCode agents 部署步骤在此步骤之前已部署),在 frontmatter 末尾、closing --- 之前,以零缩进的顶层字段插入 model:(不要插进 permission: 等多行 map 的缩进块内部)。值含 YAML 特殊字符时加引号,确保不破坏 frontmatter:

yaml
---
description: ...
mode: subagent
permission:
  read: allow
  edit: deny
steps: 12
model: provider/model-id
---
  • 如果 agent 文件已有 model: 字段(重新部署场景),替换该顶层 model: 的值,不新增重复键
  • 保留现有模型:写回 Step 0 缓存的该 agent model
  • 跳过,用主模型:不写入 model: 字段
  • 检测失败/超时、没走到本步骤的等级:用 Step 0 缓存回填 model:,避免 replace 抹掉用户上次配置
2.5 部署 Session State 模板
  • 读取 skills/story-setup/references/templates/上下文.md.tmpl
  • 仅当已识别为长篇书目且 {书名}/追踪/ 已存在时,创建缺失的 {书名}/追踪/上下文.md
  • 如果目标文件已存在,不覆盖;短篇项目不得因此创建 追踪/ 目录
2.6 合并 Hooks 注册到 settings.local.json

兼容性说明:settings-hooks.json 中 PreToolUse 的 if 字段使用 Claude Code hook 条件语法,需要运行环境支持 hook-level if。若目标工具不支持该字段,hook 脚本本身仍会自检并 advisory-only 退出;部署时可删除该 if 字段并保留 matcher + command。

  • 读取 skills/story-setup/references/templates/settings-hooks.json
  • 读取用户项目的 .claude/settings.local.json(如存在)
  • 合并 hooks 配置(按「settings-hooks.json 合并算法」处理)
  • 写入 .claude/settings.local.json

Codex hooks.json 合并算法(target_cli 含 codex 时)

Codex 项目 hooks 部署到 .codex/hooks.json,hook 脚本部署到 .codex/hooks/story_codex_hook.py。

  1. 读取 skills/story-setup/references/codex/hooks/hooks.json
  2. 读取用户现有 .codex/hooks.json(如存在),提取 hooks 部分
  3. 对每个 hook event(SessionStart、PreToolUse、PreCompact、PostCompact、Stop)按 command 去重追加;每个 hook 同时携带 command(POSIX sh,Unix)与 commandWindows(cmd.exe,Windows)两个字段,整体保留不要拆开
  4. 保留用户已有其他 hooks/config,不覆盖未知字段
  5. 写入 .codex/hooks.json 后提示用户:项目 .codex/ 层需要被 Codex trust,非 managed command hooks 还需要在 /hooks 中 review/trust 后才会运行;Windows 下 Codex 以 cmd.exe 跑 hook,走 commandWindows(cwd 为项目根时生效,否则 no-op),正文守卫在 Windows 非项目根目录下不强制拦截
Show full SKILL.md (715 more words)Show less

OpenClaw skills-only 部署算法(target_cli 含 openclaw 时)

OpenClaw Phase 1 只部署 skills,不部署 OpenClaw agents/hooks/plugin。

  1. 读取仓库当前 skills/ 下所有包含 SKILL.md 的 story skill 目录(8 个:story*)。
  2. 写入目标项目 skills/{skill-name}/,仅替换这些 story-setup 管理的已知 skill 目录;保留用户在 skills/ 下的其他目录。
  3. 每个 SKILL.md 必须满足 OpenClaw frontmatter 约束:name / description 是单行键值,metadata 是单行 JSON 对象且含 metadata.openclaw。
  4. 复制 skills/story-setup/references/openclaw/AGENTS.md.tmpl 到项目 AGENTS.md,按「AGENTS.md 合并策略」合并。
  5. .story-deployed 的 target_cli 写入 openclaw 或多端组合;references_dir 对 OpenClaw 写 skills/story-setup/references/agent-references。
  6. 安装报告必须提示:OpenClaw 会在 session 启动时 snapshot eligible skills;部署后如果命令/skills 未出现,需新开 OpenClaw session 或等待 skills watcher 刷新。
  7. 安装报告必须提示:OpenClaw Phase 1 没有硬 hooks/agents;写正文前大纲守卫、commit 提醒、session/compact 自动注入只作为 skill 内软约束,不是运行时强制拦截。
2.7 创建部署标记
  • 创建 .story-deployed 文件(sentinel file)
  • 写入以下字段(YAML key: value 格式,hook 用 references/templates/hooks/lib/sentinel.sh 读取):
    deployed_at: <date -u +"%Y-%m-%dT%H:%M:%SZ">
    agents_version: 16
    setup_skill_version: 1.2.5
    target_cli: claude-code(或 opencode、codex、openclaw,或 claude-code,opencode,codex,openclaw 等组合)
    resolver_strategy: project-local-skill-reference
    references_dir: .claude/skills/story-setup/references/agent-references(Codex 可写 .codex/skills/story-setup/references/agent-references;OpenClaw 可写 skills/story-setup/references/agent-references;多端用逗号分隔)
  • 此文件供 session-start.sh 和写作 skill 检测部署状态,避免重复提示
  • 同时创建一次性标记文件 .claude/.agents-pending-restart(空文件即可)。session-start.sh 在下一个会话启动时据此确认 agents 已随新会话注册,并自动删除该标记——用来向用户确认「重启已生效」。
  • 如果 .story-deployed 已存在但无 agents_version 或版本 < 16,提示用户重新运行 story-setup 以更新 hooks/agents/rules/reference bundle(具体变更见 UPGRADING.md)

Phase 3:验证安装

  1. 验证 hooks 注册:
    • 检查 .claude/settings.local.json 中的 hooks 字段是否正确
    • 检查 .claude/hooks/ 下的脚本是否存在且有执行权限
    • 检查 .claude/hooks/lib/common.sh 与 .claude/hooks/lib/sentinel.sh 是否存在
  2. 验证 rules 路径:
    • 检查 .claude/rules/ 下的规则文件是否存在且包含 paths frontmatter
  3. 验证 agents:
    • 检查 .claude/agents/ 下的 7 个 agent 定义文件是否存在
  4. 验证 agent reference bundle:
    • 检查 .claude/skills/story-setup/references/agent-references/ 下 reference 文件完整
    • 检查所有 story-setup/references/agent-references/<file>.md 都能解析到 deployed bundle
  5. 验证部署标记:
    • 检查 .story-deployed 是否存在且包含时间戳、agents_version: 16、setup_skill_version: 1.2.5、target_cli、resolver_strategy、references_dir
  6. 输出安装报告:
    • 列出所有已部署的文件
    • 列出需要注意的事项(如已有配置已合并)
    • ⚠️ 重启提示(必须醒目输出):本次部署写入了 .claude/agents/,但这些 custom agent 只在「会话启动」时才会被 Claude Code 注册成 subagent_type。请新开一个 Claude Code 会话再开始写作,否则当前会话里 story-review / story-long-write 等想 spawn story-architect、narrative-writer 等时会拿到「subagent_type 不可用」并降级 solo(单视角,失去多 agent 协作)。判断是否生效:新会话里跑 /story-review,报告头若是 Effective Mode: full/lean 即注册成功;若是 Fallback: ... -> solo 说明还在旧会话或未注册。
    • 重启后即可使用 /story-long-write 或 /story-short-write
    • 如果执行了 2.4.4 模型配置,输出 Agent 模型配置摘要:
      Agent 模型配置:
        story-architect          → <高端模型>(provider/model-id)
        narrative-writer         → <中端模型>(provider/model-id)
        character-designer       → <中端模型>(provider/model-id)
        story-researcher         → <中端模型>(provider/model-id)
        consistency-checker      → <低端模型>(provider/model-id)
        story-explorer           → <低端模型>(provider/model-id)
    • 如果自动检测失败(opencode models 不可用),输出手动配置指南:
      无法自动检测模型列表。以下 Agent 未配置模型,将使用主模型,成本可能较高:
        - consistency-checker(建议使用低成本模型)
        - story-explorer(建议使用低成本模型)
      
      手动配置方法:编辑 .opencode/agents/{agent名}.md,在 frontmatter 中添加:
        model: provider/model-id
      
      可用模型列表与成本可通过 opencode models --verbose 查看(输出含每模型 cost/context)。
      模型库与定价见 OpenCode 官方模型源 https://models.dev/。
  7. 验证 opencode 部署(仅当 target_cli 含 opencode 时):
    • 检查 .opencode/agents/ 下的 7 个 agent 定义文件是否存在,且 frontmatter 包含 mode: subagent 和 permission 字段
    • 检查 .opencode/plugins/story-hooks.ts 是否存在
    • 检查 .opencode/commands/ 下的 7 个 command 文件是否存在
    • 检查 skills/story-setup/references/agent-references/ 下 reference 文件完整且数量与源目录一致
    • 检查 opencode.json 的 plugin 数组是否包含 story-hooks 条目
    • 检查 .git/hooks/pre-commit 是否存在且有执行权限(Windows 上跳过执行权限检查)
    • 检查 .opencode/agents/ 下 agent 文件 frontmatter 可被 YAML 解析、model:(如有配置)是合法顶层标量,而非仅 grep 到 model: 子串
  8. 验证 Codex 部署(仅当 target_cli 含 codex 时):
    • 检查 AGENTS.md 含 Codex story skill routing sections
    • 检查 .codex/agents/ 下 7 个 .toml agent 定义文件存在并可解析
    • 检查 .codex/hooks.json 存在且 JSON 有效,包含 .codex/hooks/story_codex_hook.py command
    • 检查 .codex/hooks/story_codex_hook.py 存在且 Python 语法有效
    • 检查 .codex/skills/story-setup/references/agent-references/ 下 reference 文件完整且数量与源目录一致
    • 安装报告必须提示:Codex 需要 trust 项目 .codex/ 配置层,并在 /hooks review/trust 非 managed hooks;部署后新开 Codex 会话让 custom agents 生效;若当前运行时仍返回 unknown agent_type,按各 skill 的 fallback 规则降级 solo/direct
  9. 验证 OpenClaw 部署(仅当 target_cli 含 openclaw 时):
    • 检查 AGENTS.md 含 OpenClaw story skill routing sections
    • 检查 skills/ 下 8 个 story skill 目录存在,且每个 SKILL.md 包含单行 name、单行 description、单行 JSON metadata.openclaw
    • 检查 skills/story-setup/references/agent-references/ 下 reference 文件完整且数量与源目录一致
    • 安装报告必须提示:OpenClaw Phase 1 是 skills-only;未部署 OpenClaw agents/hooks,运行时硬拦截不可用;部署后新开 OpenClaw session 或等待 watcher 刷新

模板占位符

占位符替换规则示例
{项目名}用户项目名称或目录名《剑来》、《暗卫》
{书名}书名目录名(与目录一致)与 {项目名} 相同,或用户自定义
{目标平台}目标发布平台起点、番茄、晋江、知乎盐言
{作者名}用户笔名或昵称未指定时用「作者」

替换时去掉花括号。如果用户未指定项目名,用当前目录名。未指定的占位符保留原样不替换。

CLAUDE.md 合并策略

用户已有 CLAUDE.md 时,按 marker/section 合并:

  1. 优先识别 story-setup 管理块标记(如果旧项目已有标记,只替换标记内内容)
  2. 无标记时,读取用户现有 CLAUDE.md,按 ## 标题切分为 section map
  3. 读取模板 CLAUDE.md.tmpl,同样切分
  4. 模板中的标准 section(Skill 路由表、文件结构、协作规则、Context Recovery、语言)覆盖用户同名 section
  5. 用户独有的 section(自定义内容)保留不动
  6. 未知冲突用 AskUserQuestion 让用户选择保留哪个版本

AGENTS.md 合并策略(OpenCode / Codex / OpenClaw)

用户已有 AGENTS.md 时,按 marker/section 合并:

  1. 优先识别 story-setup 管理块标记(如果旧项目已有标记,只替换标记内内容)
  2. 无标记时,读取用户现有 AGENTS.md,按 ## 标题切分为 section map
  3. OpenCode 使用 skills/story-setup/references/opencode/AGENTS.md.tmpl;Codex 使用 skills/story-setup/references/codex/AGENTS.md.tmpl;OpenClaw 使用 skills/story-setup/references/openclaw/AGENTS.md.tmpl
  4. 模板中的标准 section(Skill 路由表、文件结构、协作规则、Compact 后恢复上下文)覆盖同名 section;用户独有 section 保留
  5. 多端同时部署时,Codex/OpenCode/OpenClaw 共同可用的通用段落只保留一份;工具特有说明以小节区分,避免互相覆盖

settings-hooks.json 合并算法

hooks 注册合并按 command 字段去重:

  1. 读取用户现有 .claude/settings.local.json(如存在),提取 hooks 部分
  2. 读取 settings-hooks.json 模板,提取要注册的 hooks
  3. 对每个 hook event(SessionStart、PreToolUse 等):
    • 用户已有的 hook command → 保留,不重复添加
    • 模板中的新 hook command → append 到对应 event 的 hooks 数组
    • 用户独有的其他配置(permissions、env 等)→ 完整保留
  4. 写入合并后的完整 settings.local.json

重新部署

  • .story-deployed 不存在 → 全新安装,Phase 2 全部执行
  • .story-deployed 存在且 agents_version: 16 → 提示已部署,AskUserQuestion 确认是否重新部署
  • .story-deployed 存在但 agents_version < 16 → 提示需要更新,重新执行 Phase 2 覆盖 agents/hooks/rules/reference bundle,CLAUDE.md / AGENTS.md / settings.local.json / .codex/hooks.json 走合并策略

参考资料

文件用途
references/templates/CLAUDE.md.tmpl项目根 CLAUDE.md 模板
references/templates/hooks/8 个 hook 脚本模板 + lib/common.sh/lib/sentinel.sh(正文兜底 check-prose-after-write.sh 限 PostToolUse Write/Edit;cat>/tee 等 Bash 写正文由 Codex Stop 回合末 git 扫描兜,Claude/OpenCode 的 Bash 仅 pre-guard)
references/templates/rules/4 条 path-scoped 规则模板
references/templates/agents/6 个 agent 定义模板(story-architect, character-designer, narrative-writer, consistency-checker, story-researcher, story-explorer)
references/agent-references/Agent 模板自带的参考资料副本;部署到 .claude/skills/story-setup/references/agent-references/,避免跨 skill references
references/templates/settings-hooks.jsonhooks 注册 JSON 片段
references/templates/上下文.md.tmpl写作上下文模板
references/codex/AGENTS.md.tmplCodex 项目根 AGENTS.md 模板
references/codex/agents/7 个 Codex custom agent TOML 模板
references/codex/hooks/hooks.jsonCodex hooks 注册 JSON 模板(部署到 .codex/hooks.json)
references/codex/hooks/story_codex_hook.pyCodex hook adapter(部署到 .codex/hooks/story_codex_hook.py)
references/openclaw/AGENTS.md.tmplOpenClaw 项目根 AGENTS.md 模板(skills-only)

流程衔接

流水线: 部署 位置: 初始化(最前置)

时机跳转到命令
部署完成,开始写作novel-pipeline / novel-write(story-long-write / story-short-write 只作方法参考)novel-studio --pipeline --prompt-file <需求文件>

© Xiaoyangy, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 91 other files (references) in skills/story-setup of Xiaoyangy/novel-studio.

  • SKILL.md
  • CONTEXT.md
  • UPGRADING.md
  • context.json
  • references/agent-references/anti-ai-writing.md
  • references/agent-references/banned-words.md
  • references/agent-references/character-basics.md
  • references/agent-references/character-design-methods.md
  • references/agent-references/character-relations.md
  • references/agent-references/dialogue-mastery.md
  • references/agent-references/emotional-arc-design.md
  • references/agent-references/emotional-methods.md
  • references/agent-references/format-and-structure.md
  • references/agent-references/genre-catalog.md
  • references/agent-references/genre-core-mechanics.md
  • references/agent-references/genre-readers.md
  • references/agent-references/genre-writing-formulas.md
  • references/agent-references/hooks-chapter.md
  • references/agent-references/hooks-suspense.md
  • … and 73 more

Open the folder on GitHubat commit ed04a10

Compare with similar skills

Story Setup next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Story Setup compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Story Setup this skillXiaoyangy/novel-studio131—~5.7kAutomated safety check: PassApache-2.0
Claude Project Setupmohitagw15856/pm-claude-skills1.4k—~1.2kAutomated safety check: PassMIT
Using Agent Skillsaddyosmani/agent-skills104k4 repos~2.4kAutomated safety check: PassMIT
Claude ReflectBayramAnnakov/claude-reflect1.8k2 repos~627Automated safety check: PassMIT
Neat-Freak Knowledge CloseoutKKKKhazix/khazix-skills21k—~1.9kAutomated safety check: PassMIT
Writing For Agentsbestofjs/bestofjs3.1k18 repos~2.7kAutomated safety check: PassMIT

Similar skills

  • Claude Project Setup

    mohitagw15856/pm-claude-skills

    Set up a repo or project so an AI coding agent works well in it — the CLAUDE.md, the context, the guardrails, and the conventions the agent needs to be useful instead of lost.

    1.4k GitHub stars~1.2k tokensUpdated yesterday
    Agent WorkflowsAuto-check passed
  • Using Agent Skills

    addyosmani/agent-skills

    Meta-skill for choosing which workflow skill fits the task at hand, plus always-on habits: surface assumptions, stop on confusion, push back, keep it simple and stay in scope.

    104k GitHub starsUsed in 4 repos~2.4k tokens
    Agent WorkflowsAuto-check passed
  • Claude Reflect

    BayramAnnakov/claude-reflect

    Self-learning system that captures corrections during sessions and reminds users to run /reflect to update CLAUDE.md.

    1.8k GitHub starsUsed in 2 repos~627 tokens
    Agent WorkflowsAuto-check passed
  • Neat-Freak Knowledge Closeout

    KKKKhazix/khazix-skills

    Brings project docs, agent rule files, authorized memory and leftover workspace files back in line with what the code and runtime actually do at the end of a work session.

    21k GitHub stars~1.9k tokensUpdated 9 days ago
    Agent WorkflowsAuto-check passed
  • Writing For Agents

    bestofjs/bestofjs

    Writing documents for agents. An agent skill from bestofjs/bestofjs.

    3.1k GitHub starsUsed in 18 repos~2.7k tokens
    Agent WorkflowsAuto-check passed
  • Task Observer

    rebelytics/one-skill-to-rule-them-all

    Monitors task execution for skill improvement opportunities.

    3.2k GitHub starsUsed in 1 repo~11k tokens
    Agent WorkflowsAuto-check passed

More from Xiaoyangy/novel-studio

All 18 skills in this repo
  • Story Long Write

    Xiaoyangy/novel-studio

    长篇网文写作方法论与 novel-studio pipeline 适配器。用于把长篇开书、大纲、日更、续写、重写等需求整理为 pipeline 输入;在 novel-studio 内禁止直接生成正文。触发方式:/story-long-write、/写长篇、「帮我开书」「写大纲」「日更」「续写」「继续写」「修改第X章」「回炉」「重写第X章」。

    131 GitHub stars~9k tokensUpdated 17 days ago
    Auto-check passed
  • Story Short Write

    Xiaoyangy/novel-studio

    短篇网文写作方法论与 novel-studio pipeline 适配器。用于把短篇、盐言故事、拆文续写等需求整理为 pipeline 输入;在 novel-studio 内禁止直接生成正文。触发方式:/story-short-write、/写短篇、「帮我写一篇短篇」「写个盐言故事」。

    131 GitHub stars~5.6k tokensUpdated 17 days ago
    Auto-check passed
  • Story Douban Long Write

    Xiaoyangy/novel-studio

    豆瓣阅读原创长篇专项方法论与 novel-studio pipeline 适配器。用于把 15W-30W 字原创长篇的主题、类型、开头、文字、标签、简介和提交自检要求整理为 pipeline 输入;在 novel-studio 内禁止直接生成正文。触发方式:/story-douban-long-write、/豆瓣长篇、/豆瓣写作、「写一部豆瓣阅读原创长篇」「按豆瓣要求写长篇」。

    131 GitHub stars~1k tokensUpdated 17 days ago
    Auto-check passed
  • Story Short Analyze

    Xiaoyangy/novel-studio

    短篇网文拆文。拆解爆款短篇小说(番茄短篇 / 故事会 / 知乎盐选 / 追妻 / 世情 / 重生 / 虐渣等通俗题材)的故事核、结构、情感线、反转设计、写作手法、共鸣层次。单一全量拆解管道:跑完 Stage 2-6 产出完整拆文报告,落盘到 deconstruction-library/{书名}/,下游 story-short-write 同时读拆文报告 + 情节节点 + 写作手法 + 原文…

    131 GitHub stars~2.3k tokensUpdated 17 days ago
    Auto-check passed
  • AI Novel Audit

    Xiaoyangy/novel-studio

    审核一段/一篇小说(或任意叙事性文本)的「AI 创作度」——判断它有多大可能由 AI 生成、或 AI 参与了多少。当用户想知道一段文字是不是 AI 写的、AI 味重不重、是人工还是 AI、AI 辅助了多少、帮忙鉴别/检测/审稿 AI 生成内容、给稿件的「人味/机味」打分时,都使用本技能。触发词包括但不限于:AI 创作度、AI 味、是不是 AI 写的、AI 生成检测、人工还是 AI、鉴别…

    131 GitHub stars~3.1k tokensUpdated 17 days ago
    Auto-check passed
  • Novel Pipeline

    Xiaoyangy/novel-studio

    把各功能串成一条可恢复的流水线:(共创→)写作→评审→重写→交付,按阶段执行、断点续跑。触发:「一条龙写完整本书」「从头跑完整个流程」「中断后接着跑」,想要无人值守端到端、且能中断恢复时使用。

    131 GitHub stars~632 tokensUpdated 17 days ago
    Auto-check passed

Categories

Questions about Story Setup

What does Story Setup do?

网文写作工具集基础设施部署。将 hooks/rules/agents/CLAUDE.md/AGENTS.md 等基础设施部署到用户项目目录,支持 Claude Code / OpenCode / Codex / OpenClaw。触发方式:/story-setup、$story-setup、「准备写书」「帮我搭一下环境」「配置写作项目」。. Story Setup is an agent skill from Xiaoyangy/novel-studio.

When should I use Story Setup?

Story Setup fits situations like: tasks that involve Agent instruction files; tasks that involve Creative writing and fiction.

How do I install Story Setup in Claude Code?

Run `npx skills add Xiaoyangy/novel-studio --skill story-setup -a claude-code`. Or copy the skill folder (skills/story-setup in Xiaoyangy/novel-studio) into .claude/skills/story-setup in your project. Claude Code loads it when a task matches its description.

How do I install Story Setup in Codex?

Run `npx skills add Xiaoyangy/novel-studio --skill story-setup -a codex`. Or copy the skill folder (skills/story-setup in Xiaoyangy/novel-studio) into .agents/skills/story-setup in your project. Codex loads it when a task matches its description.

Can I use Story Setup in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add Xiaoyangy/novel-studio --skill story-setup -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/story-setup, .gemini/skills/story-setup, .github/skills/story-setup and .opencode/skills/story-setup in your project.

What does Story Setup need to run?

Going by SKILL.md and its folder, Story Setup needs the command-line tools its instructions call (opencode). Our summary lists: Python 3.

Does Story Setup access the network?

SKILL.md names 1 domain. As links in the text: models.dev. This is read from the text; nothing was executed.

Is Story Setup safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Story Setup use?

Story Setup is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Story Setup use?

About 5.7k tokens (SKILL.md is roughly 23k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 196k tokens, read only when the agent opens those files.

What are the alternatives to Story Setup?

Skills that share tags, products or a category with Story Setup: Claude Project Setup (mohitagw15856/pm-claude-skills, 1.4k stars), Using Agent Skills (addyosmani/agent-skills, 104k stars), Claude Reflect (BayramAnnakov/claude-reflect, 1.8k stars) and Neat-Freak Knowledge Closeout (KKKKhazix/khazix-skills, 21k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Story Setup?

Xiaoyangy (a GitHub user) maintains it in Xiaoyangy/novel-studio, which has 131 GitHub stars. The repository holds 18 skills in this directory. The repository was last updated on September 23, 2026.

Source: Xiaoyangy/novel-studio on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.