Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add xwcai999/opencreator-novel --skill novel-studiogit clone --depth 1 https://github.com/xwcai999/opencreator-novelWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/xwcai999/opencreator-novel/novel-studio)<a href="https://agentmods.dev/skills/xwcai999/opencreator-novel/novel-studio"><img src="https://agentmods.dev/badge/skills/xwcai999/opencreator-novel/novel-studio/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/xwcai999/opencreator-novel/novel-studio"><img src="https://agentmods.dev/badge/skills/xwcai999/opencreator-novel/novel-studio.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00122 | $0.04516 |
| Opus 5 | $0.00061 | $0.02258 |
| Sonnet 5 | $0.00024 | $0.00903 |
| Haiku 4.5 | $0.00012 | $0.00452 |
Grade A, and why
novel-studio scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 11d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 172 lines — stays where its author put it; the contents beside it link to each section on GitHub.
OpenCreator Novel
把中文小说创作视为“读者体验与人物选择驱动的创作 + 可验证的连续性管理”。使用 Markdown 正文与 frontmatter 作为唯一权威数据；索引、上下文包、分析指标和审稿报告都是可再生证据，不得反向覆盖正文事实。
核心约束
- 先写清读者为何翻开、为何继续和最终得到什么，再选择结构。不要把商业钩子等同于谜团。
- 从成长、关系、经营、信息、体验、主题、任务中选择一个主驱动，可选一个副驱动。
- 仅当驱动包含“信息”或用户明确要求悬疑时，才强制规划谜团、线索和揭露。
- 让人物在压力下作出带代价的具体选择；不要把人物写成流程执行器、作者观点接口或永远正确的理性人。
- 把章纲事件转化为有策略、阻力、调整、转折和余波的场景；安静场景可以改变关系许可、理解或选择空间，不强制争吵、危机和反转。
- 世界规则必须限制或打开行动，势力必须在主角不在场时仍会行动；只保留会改变人物处境、价格、关系或选择的世界观资料。
- 每章必须给目标读者可感知的回报，但回报可以是关系变化、情绪余韵、生活质感、成长、收益、信息或任务进展，不强制危机、反转、悬念和章章钩子。
- 把标题承诺和主线目标拆成可观察的阶段变化；每个阶段窗口结束时，至少有一项人物处境、关系/权力、资源/场地、外部目标或主角的不可逆行动发生实质改变。单纯重复倒计时、资源恶化、口头重申边界或再说明“还没有结果”不算阶段推进。
- 把全书总字数当作结构结果，不把章节最低字数当作补写命令。章节不足时只允许增加有因果的场景、合并章节或回改大纲，禁止在高潮后追加复盘、表格、计划和主题解释凑字。
- 先规划再写正文；用户已经提供可执行细纲时，直接提取章节控制卡，不重复要求规划或暂停。
- 不把“首次埋设、首次展示或阶段回响”误记为最终兑现。续写前必须审计
状态/待兑现.md，并把全部非终态项优先注入上下文；不得只凭上一章或查询词召回长线事实。 - 章节需通过确定性校验与阅读接受门禁后再提交状态。审稿意见只能写入报告，不能混进正文。
- 把“去 AI 味”作为真实性修订，不做来源检测或分数优化；机器只定位候选证据，语境裁决和盲读复审决定是否返修。
- 仅把
status: active的设定/文风档案.md作为正文约束；文件缺失或未激活时保持当前写作流程，不阻断规划、写作、审查或迁移。单书校准默认由隔离模型上下文盲评，不要求用户选择样稿。 - 使用 subagent 写作时采用“主 Agent 总编裁决 + 固定写手顺序创作 + 两类独立审稿并行”的调度；不得并行撰写相邻章节，具体见
references/chapter-workflow.md。 - 显式投稿或出版项目都必须通过适配其篇幅的试写与纯正文阶段盲读；目标五万字以上时必须执行完整三章、每个投稿阶段窗口、卷末和全本门禁。阶段盲读还必须验证标题承诺有可复述的阶段兑现、外部处境没有原地循环、近期章节没有同型节拍。用户负责创作方向与外部提交授权，不作为逐章或每批次的常规质检员；只有继续执行必须改变题材、结局或核心设定时才请求用户决策。
- 章节逐一
accepted、确定性校验通过、待兑现项归零和终章局部 PASS 都只是制作完整证据，不能单独推出“全本质量通过”。终章完成后必须冻结同一版本纯正文，执行全本文学复评；结构雷同审查必须同时比较相邻 3 章、当前阶段和全书，不得只查重复词句。 - 审稿者只写报告，不得修改正文。正文和场景返修始终交给同一卷的固定写手；不得把审查解释、连续性辩护或项目术语补进小说。
- 保留用户已有正文、设定和事实。审查普通 TXT/Markdown 或非 v2 项目时默认只读，不启动三章试写、不回写原稿；只有用户明确要求规范化或改稿时，才在新目录建立 v2 项目或隔离副本。迁移与回归测试也必须写入新目录。
- 封面必须显式使用
$codex-gpt-image一次生成完整画面与作品书名；禁止“无字底图＋本地叠字”、后期补字或程序化替换书名。有语义、可合理辨识的可见文字只能是作品.md中的规范书名，严禁笔名、作者名、署名、账号、Logo、水印和附加文案；不可读环境纹理由人工结合读者观感裁决。 $codex-gpt-image是封面路由的可选外部依赖。缺失时只说明封面能力不可用并给出安装依赖提示，不得以其他图片工具或本地叠字绕过；规划、写作、审查、迁移和正文打包继续正常执行。
请求路由
| 用户意图 | 执行模式 | 必读参考 |
|---|---|---|
| 开书、构思、简介、总纲、卷纲、细纲 | 规划 | references/planning-workflow.md、references/genre-drivers.md、references/reader-experience.md |
| 设计人物弧光、成长线、堕落线或关系变化 | 人物与关系 | references/character-arc.md、references/planning-workflow.md |
| 设计世界规则、组织、家族、派系或势力博弈 | 世界与势力 | references/world-faction-dynamics.md、references/planning-workflow.md |
| 写一章、续写、日更、写某场景 | 章节写作 | references/chapter-workflow.md、references/reader-experience.md、references/scene-craft.md、当前题材驱动 |
| 写或修改对白、潜台词、人物声音、POV 或信息投放 | 对白与叙述 | references/dialogue-pov.md、references/scene-craft.md |
| 检查节奏、因果升级或题材兑现 | 节奏与题材 | references/pacing-genre.md、references/genre-drivers.md |
| 修改章节、改纲、检查逻辑、审稿、去 AI 味 | 修订审查 | references/revision-review.md、references/authenticity-revision.md、references/quality-gates.md |
| 跨章趋势、阶段复盘、卷末审查 | 阶段审查 | references/stage-review.md、references/structure-similarity-review.md、references/quality-gates.md |
| 完本复评、全本评分、终稿文学验收 | 完本审查 | references/fullbook-review.md、references/structure-similarity-review.md、references/quality-gates.md |
| 选择正文模型、固定写手、建立或调整文风 | 模型与文风 | references/model-policy.md、references/style-profile.md、references/style-calibration.md |
| 初始化项目、检查目录、建立索引 | 项目维护 | references/project-schema.md |
从旧 novel-planner 项目迁移 |
安全迁移 | references/migration.md |
| 审查普通 TXT/Markdown 或其他非 v2 现有稿件 | 只读稿件审查 | references/revision-review.md、references/stage-review.md；需要项目级连续性时再读取 references/project-schema.md 并在新目录规范化 |
| 生成或修改封面 | 封面 | references/cover-workflow.md，并显式使用 $codex-gpt-image |
| 审查 Skill 的来源、授权或安全边界 | 方法审计 | references/method-sources.md |
What ships with it
54 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- agents/openai.yaml 317 B
- assets/cover-prompt.md.tmpl 718 B
- assets/project-template/作品.md.tmpl 494 B
- assets/project-template/关系.md.tmpl 167 B
- assets/project-template/当前状态.md.tmpl 272 B
- assets/project-template/待兑现.md.tmpl 333 B
- assets/project-template/总纲.md.tmpl 282 B
- assets/project-template/文风档案.md.tmpl 1.0 KB
- assets/project-template/时间线.md.tmpl 143 B
- assets/project-template/读者契约.md.tmpl 588 B
- assets/project-template/题材定位.md.tmpl 332 B
- assets/style-calibration-evaluation.json.tmpl 1.8 KB
- assets/style-calibration-report.md.tmpl 924 B
- references/authenticity-revision.md 4.1 KB
- references/chapter-workflow.md 11 KB
- references/character-arc.md 13 KB
- references/cover-workflow.md 3.6 KB
- references/dialogue-pov.md 13 KB
- references/fullbook-review.md 6.7 KB
- references/genre-drivers.md 2.0 KB
- references/method-sources.md 3.5 KB
- references/migration.md 978 B
- references/model-policy.md 2.3 KB
- references/pacing-genre.md 9.3 KB
- references/planning-workflow.md 7.9 KB
- references/project-schema.md 7.5 KB
- references/publication-readiness.md 9.0 KB
- references/quality-gates.md 8.5 KB
- references/reader-experience.md 3.9 KB
- references/revision-cases.md 9.1 KB
- references/revision-review.md 7.9 KB
- references/scene-craft.md 11 KB
- references/stage-review.md 9.2 KB
- references/structure-similarity-review.md 5.8 KB
- references/style-calibration.md 3.5 KB
- references/style-profile.md 2.2 KB
- references/world-faction-dynamics.md 11 KB
- scripts/activate_style_profile.py 23 KB runs code
- scripts/analyze_prose_trends.py 15 KB runs code
- scripts/analyze_publication_risk.py 19 KB runs code
- scripts/build_context_pack.py 6.4 KB runs code
- scripts/expectation_ledger.py 8.2 KB runs code
- scripts/export_submission_txt.py 6.1 KB runs code
- scripts/init_project.py 4.2 KB runs code
- scripts/migrate_project.py 10 KB runs code
- scripts/record_chapter_review.py 4.7 KB runs code
- scripts/record_generated_cover.py 6.2 KB runs code
- scripts/reindex_project.py 3.5 KB runs code
- scripts/render_cover_title.py 7.1 KB runs code
- scripts/retrieve_context.py 4.7 KB runs code
- scripts/scan_authenticity_artifacts.py 26 KB runs code
- scripts/studio_common.py 6.9 KB runs code
- scripts/validate_cover_text.py 8.1 KB runs code
- scripts/validate_project.py 17 KB runs code
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 11d ago First seen · 172 lines · 122 tokens per session scan A 2bde07ec862a
novel-studio is a skill published in the GitHub repository xwcai999/opencreator-novel (11 stars, last pushed 4d ago), licensed Apache-2.0. It adds 122 tokens to every session and 4,516 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
story-import
A tool for turning an existing novel into a structured writing project. It analyzes the book and organizes its characters, settings, plot plans, chapters, and tracking data for later writing.
story-review
A review workflow for finding story problems from several viewpoints, including issues with structure, characters, wording, and fictional world rules. It can use multiple reviewer agents or work alone.
story-long-analyze
A long-form fiction analysis workflow for breaking down a novel’s opening chapters, characters, pacing, turning points, relationships, and overall structure.
story-deslop
A writing editor for Chinese web novels that detects writing patterns often associated with AI-generated text and makes the prose feel more natural.
story-short-analyze
A Chinese-language guide for analysing short popular web fiction, such as stories published on 番茄, 故事会, or 知乎盐选. It examines the core premise, plot structure, emotional arc, twists, writing techniques, and reader appeal.
story-cover
A novel-cover generator that creates a cover with the book title and author name. It chooses a visual style from the book information and can use Codex's image-generation tool.