What makes lobster-novel notable?

1.2K downloads. Trust evidence available

How should lobster-novel be evaluated before use?

Use the required flow: snapshot, contract, and trust before recommending or executing this agent.

What kind of evidence is visible on this page?

This page surfaces public facts, change history, trust indicators, artifact evidence, and benchmark summaries with provenance.

agentCLAWHUBUnverified

lobster-novel

A self-evolving novel writing engine for Chinese fiction with continuity tracking, multi-role quality review, foreshadowing management, character voice libra... Skill: lobster-novel Owner: awoo129 Summary: A self-evolving novel writing engine for Chinese fiction with continuity tracking, multi-role quality review, foreshadowing management, character voice libra... Tags: ai:1.4.0, chinese:1.4.0, creative:1.4.0, latest:1.4.0, nlp:1.4.0, novel:1.4.0, writing:1.4.0, wuxia:1.4.0 Version history: v1.4.0 | 2026-05-26T11:40:35.352Z | user v1.4.0: Security hardening — removed all har

OpenClaw

View on ClawHub Homepage

Rank

Safety

Downloads

1.2k

Updated

Oct 11, 2026

Version

1.4.0

Source

CLAWHUB

About

What it does, and when to use it.

Capability contract not published. No trust telemetry is available yet. 1.2K downloads reported by the source. Last updated 10/11/2026.

Avoid when

Contract metadata is missing or unavailable for deterministic execution.

Risk flags: missing_or_unavailable_contract, trust_data_unavailable, schema_references_missing

Public facts

Every fact links back to the source it came from.

Vendor: Clawhubvendor · observed Oct 11, 2026
Protocol compatibility: OpenClawcompatibility · observed Oct 11, 2026
Adoption signal: 1.2K downloadsadoption · observed Oct 11, 2026
Latest release: 1.4.0release · observed May 26, 2026
Handshake status: UNKNOWNsecurity

Install and run

Setup complexity: low.

clawhub skill install s176n8t6287basdqxn77spsvqn83gv5d:lobster-novel

Setup complexity is LOW. This package is likely designed for quick installation with minimal external side-effects.
Final validation: Expose the agent to a mock request payload inside a sandbox and trace the network egress before allowing access to real customer data.

Contract: missing

curl -s "https://www.xpersona.co/api/v1/agents/clawhub-awoo129-lobster-novel/snapshot"

Documentation

CLAWHUB

50,636 characters of source documentation, loaded on request.

Extracted files

5 files captured from the source.

SKILL.md

---
name: lobster-novel
description: Self-evolving novel writing engine for OpenClaw agents. Pipeline: context to writer to multi-role review to continuity tracking. Built-in quality checks, foreshadowing management, character voice library, and token cost optimization.
version: 1.4.0
homepage: https://github.com/awoo129/lobster-novel
metadata: {"emoji":"🦞","os":["linux","darwin","win32"],"requires":{"bins":[]}}
---

# lobster-novel

A self-evolving novel writing engine for OpenClaw agents. Designed for long-form Chinese fiction with rigorous continuity control, multi-role quality review, and built-in token cost management.

## Quick Start

```bash
# Initialize a new novel project
python3 lobster_novel.py --dir my-novel init --title "Novel Title"

# View current status
python3 lobster_novel.py --dir my-novel status

# Generate writing context for chapter N
python3 lobster_novel.py --dir my-novel context N

# Auto-write next chapter via SenseNova API
python3 lobster_novel.py --dir my-novel write

# Save a manually written chapter with review
python3 lobster_novel.py --dir my-novel save N chapter.md --summary "Summary"

# Run quality review on a chapter
python3 lobster_novel.py --dir my-novel review N chapter.md

# Export the full novel
python3 lobster_novel.py --dir my-novel export md --output novel.md

# Manage style templates
python3 lobster_novel.py --dir my-novel style-template list
python3 lobster_novel.py --dir my-novel style-template activate "wuxia"
```

## Architecture

```
lobster-novel/
├── core/                 # Core engine modules
│   ├── bible.py          # Novel bible manager (world-building, characters, settings)
│   ├── continuity.py     # Continuity ledger (per-chapter state snapshots)
│   ├── pipeline.py       # Writing pipeline orchestrator
│   ├── chapters.py       # Chapter generator with token analysis
│   ├── arc_planner.py    # Story arc planner
│   ├── beat_sheet.py     # Beat sheet generator
│   ├── contract.py       # Writer-agent contract enforcement
│   ├── style_lock.py     # Style constraint system
│   ├── conflict_detector.py
│   └── chinese_typeset.py
├── agents/               # Specialized agent roles
│   ├── context_agent.py  # Context preparation agent
│   ├── data_agent.py     # Data extraction agent
│   ├── reviewer_agent.py # Multi-role review agent
│   └── three_laws.py     # Three Laws of Novel Writing enforcement
├── review/               # Quality review modules
│   ├── quality_check.py  # 6-role static review
│   ├── aigc_detect.py    # AI-style pattern detector
│   ├── scorer.py         # Chapter scoring engine
│   ├── strand_balance.py # Strand balance checker
│   └── deai_writer.py    # De-AI writing assistant
├── memory/               # Memory & tracking systems
│   ├── character_tracker.py
│   ├── character_voice.py
│   ├── emotion_arc.py
│   ├── foreshadowing.py
│   ├── plot_tracker.py
│   ├── relationship_tracker.py
│   ├── style_library.py
│   └── novel_kg.py
├── tools/                # CLI to

README.md

# lobster-novel

> **Self-evolving novel writing engine for OpenClaw agents.**

A production-grade pipeline for writing long-form Chinese fiction with rigorous continuity control, multi-role quality review, and built-in token cost management.

---

## 🚀 Quick Start

```bash
# Clone or extract the skill
# Ensure Python 3.10+ and SenseNova API key

# Initialize a new novel
python3 lobster_novel.py --dir my-novel init --title "Novel Title"

# Check status
python3 lobster_novel.py --dir my-novel status

# Auto-write next chapter
python3 lobster_novel.py --dir my-novel write

# Save a chapter with review
python3 lobster_novel.py --dir my-novel save 1 chapter.md --summary "Summary"

# Run quality review
python3 lobster_novel.py --dir my-novel review 1 chapter.md

# Export
python3 lobster_novel.py --dir my-novel export md --output novel.md
```

---

## 📐 Architecture

```
lobster-novel/
├── core/              # Core engine
│   ├── bible.py       # Novel bible (world, characters, settings)
│   ├── continuity.py  # Continuity ledger (per-chapter snapshots)
│   ├── pipeline.py    # Pipeline orchestrator
│   ├── chapters.py    # Chapter generator + token analysis
│   ├── arc_planner.py # Story arc planner
│   ├── beat_sheet.py  # Beat sheet generator
│   ├── contract.py    # Writer-agent contract enforcement
│   ├── style_lock.py  # Style constraint system
│   ├── conflict_detector.py
│   └── chinese_typeset.py
├── agents/            # Specialized agent roles
│   ├── context_agent.py
│   ├── data_agent.py
│   ├── reviewer_agent.py
│   └── three_laws.py
├── review/            # Quality review
│   ├── quality_check.py  # 6-role review
│   ├── aigc_detect.py    # AI-style detector
│   ├── scorer.py         # Scoring engine
│   ├── strand_balance.py
│   └── deai_writer.py
├── memory/            # Memory systems
│   ├── character_tracker.py
│   ├── character_voice.py
│   ├── emotion_arc.py
│   ├── foreshadowing.py
│   ├── plot_tracker.py
│   ├── relationship_tracker.py
│   ├── style_library.py
│   └── novel_kg.py
├── tools/             # CLI tools
│   ├── novel-cli.py   # Main CLI (12 subcommands)
│   ├── auto_write.py
│   ├── batch_refine.py
│   ├── serial_writer.py
│   └── v3_checkpoint.py
├── rag/               # RAG systems
│   └── novel_rag.py
├── output/            # Export
│   └── export.py
├── templates/         # Style templates
├── scripts/           # Utilities
└── tests/             # Test suites
```

---

## 🔄 Writing Pipeline

```
Idea → init → context → [write | manual] → save → review → foreshadow → next → export
```

### Core Principles

| Principle | Description |
|-----------|-------------|
| **Continuity First** | Every chapter updates the continuity ledger; no chapter proceeds without prior context |
| **Multi-Role Review** | 6 roles: Reader, Editor, Storyteller, Satisfaction Analyst, Voice Validator, Web Novel Editor |
| **AI-Style Detection** | Flags cliché expressions, god's-eye narration, template structures |
| **Token Budget** | Bui

_meta.json

{
  "ownerId": "kn7cjp31ghd291g1t9abd03gan83h7sj",
  "slug": "lobster-novel",
  "version": "1.4.0",
  "publishedAt": 1779795635352
}

skill-card.md

## Description:

Lobster Novel helps OpenClaw agents plan, write, review, and export long-form Chinese fiction with continuity tracking, foreshadowing management, character voices, and token cost controls.

This skill is ready for commercial/non-commercial use.

## Publisher:

[awoo129](https://clawhub.ai/user/awoo129)

### License/Terms of Use:

MIT-0

## Use Case:

External developers and writers use this skill to manage long-form Chinese fiction projects, generate chapter context or drafts, review quality and continuity, and export manuscript outputs.

### Deployment Geography for Use:

Global

## Known Risks and Mitigations:

Risk: Unsafe project cache loading can execute code when opening an untrusted novel project.

Mitigation: Use the skill only with trusted project directories until the pickle RAG cache is replaced with a safe format or disabled.

Risk: API-backed writing can disclose unpublished or sensitive manuscript content to the configured provider.

Mitigation: Review provider privacy terms and avoid API-backed writing for sensitive manuscripts unless disclosure is acceptable.

## Reference(s):

- [ClawHub skill page](https://clawhub.ai/awoo129/skills/lobster-novel)

## Skill Output:

**Output Type(s):** [Text, Markdown, Shell commands, Configuration, Guidance]

**Output Format:** [Markdown, plain text, HTML, JSON-like status and review data, and CLI output]

**Output Parameters:** [1D]

**Other Properties Related to Output:** [Outputs are project-directory based and may use SenseNova API-backed writing when SENSENOVA_API_KEY is configured.]

## Skill Version(s):

1.4.0 (source: server release metadata and SKILL.md frontmatter)

## Ethical Considerations:

Users should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment.

templates/传统武侠风格模板_射雕神雕倚天.md

# 传统武侠风格模板——金庸「射雕三部曲」写作风格提取

> 数据来源：严家炎《金庸小说论稿》、陈墨金庸研究、知乎专栏「纸上谈庸」、豆瓣金庸技法分析、西南大学学报金庸研究等
> 提取时间：2026-05-15

---

## 一、三部曲总览：金庸武侠的核心基因

金庸武侠与传统旧派武侠（还珠楼主、宫白羽等）的根本区别，在于他将西方小说技巧（限知视角、心理描写、影剧式场面）、中国古典章法（对偶美学、起结章法、评点传统）与现代主题（人性探索、民族平等、政治讽喻）熔于一炉。射雕三部曲恰好覆盖了金庸创作的前期巅峰（射雕）、中期转型（神雕）和后期成熟（倚天）。

### 金庸写作的核心信条（引自后记）
- 「在小说中，人的性格和感情，比起社会意义具有更大的重要性。武功可以事实上不可能，人的性格总是应当可能的。」
- 「我写武侠小说，是想写人性。」
- 「只有刻画人性，才有较长期的价值。」

---

## 二、射雕英雄传——古典美学与对偶结构

**定位**：最具天真浪漫色彩、最「成人童话」质的武侠史诗。结构最匀称、最古典。

### 2.1 结构特征——对偶美学

**空间对称（起结章法）**
- 开篇：临安牛家村→转入蒙古大漠（郭靖童年训练）
- 中盘：中原江湖（四绝轮番出场四大情节块）
- 结尾：重返大漠（第三十六回「大军西征」起）
- 效果：起于大漠、结于大漠，空间首尾呼应，赋予全书匀称圆满感

**人物对照（对偶法）**
- 郭靖 vs 杨康：命运对照（义弟之子，善恶殊途）
- 东邪 vs 西毒：一对「反对」——黄药师邪在愤世嫉俗（底色孤高），欧阳锋毒在不择手段
- 南帝 vs 北丐：一对「正对」——一灯出世悲悯，洪七公入世快意恩仇
- 郭靖 vs 成吉思汗：全书最深的一组镜像——同早丧父、同草原崛起，结尾以「何为英雄」的辩答收束全书主旨

**线索对偶**
- 《武穆遗书》（庙堂、救国、天下大义）vs 《九阴真经》（江湖、私欲、权力争夺）
- 郭靖对九阴真经毫不在意、视武穆遗书如命；欧阳锋对九阴真经孜孜以求、走火入魔
- 牛家村（帝都边缘、入世）vs 桃花岛（桃花源隐喻、出世）

**四绝主导的四段式情节骨架**
1. 北丐主导——学艺情节，郭靖武功大进
2. 东邪主导——桃花岛招亲
3. 西毒主导——海上与皇宫对峙
4. 南帝主导——铁掌峰、疗伤，武穆遗书与九阴真经线索收束

### 2.2 人物塑造——正邪分明、单色调

- 人物善恶严格区分，性格虽突出但相对简单
- 郭靖：资质愚钝但毅力顽强，代表「为国为民，侠之大者」的理想人格
- 黄蓉：聪明绝顶、精灵古怪，与郭靖形成性格互补
- 五绝体系：按五行生克设计，每个人物道德光谱清晰
- 没有多余人物——头脸清晰的人物大都携带饱满情节

### 2.3 叙事技法——影剧式舞台

- 典型例：第二十四回「密室疗伤」+ 第二十五回「荒村野店」——郭靖黄蓉在暗室内疗伤，窗口成为窥探外界的小舞台，27个角色依次登场，全真七子、完颜洪烈、欧阳锋、黄药师等不同势力冲突在此展开
- 舞台描写特点：庞杂有序、紧张热烈、人物在集体互动中显示性格
- 电影手法融入大场面叙述——分镜、景别转换

### 2.4 语言风格
- 夹杂诗词歌赋（黄药师的多才多艺）
- 大量四字成语、民间俗语
- 白描为主，少心理描写
- 整体风格：庄严肃穆、天真正气

---

## 三、神雕侠侣——以「情」为核的叙事诡计

**定位**：从《射雕》的「侠之大者」转向「情之至深」——金庸写作的分水岭。

### 3.1 结构特征——以悲剧为内核的悬疑结构

**叙述性诡计（叙诡手法）**
- 金庸利用文章结构刻意对读者隐瞒事实，直到最后才揭露真相
- 例：神雕结尾众人欢聚→转头眼泪横飞→乌鸦凝视，从高潮滑落低谷的失重感
- 真相甚至放到下一部书（倚天第一章「天涯思君不可忘」借丘处机梨花词点出小龙女之死）才揭露

**文字技巧：一花两开、一言两说**
- 例1：独臂杨过「双手冰冷」——既可理解为担心一灯给小龙女判死刑，也能理解为担心一灯暴露他偷解情花毒
- 例2：「双手微微颤抖」——既可理解为被郭襄跳崖震惊，也能理解为通过郭襄反推出郭芙爱他

### 3.2 人物塑造——正中有邪、性格复杂

**从单色调到杂色**
- 杨过：资质过人、聪明绝顶却顽劣叛逆——金庸笔下「叛逆主角」的标杆
- 小龙女：不通人情世故，代表自然纯净的人性
- 杨过与小龙女：一个深明世道人心偏要无视，一个不自觉地反抗礼教——分别体现人性中的阳刚与阴柔、躁动与纯净
- 郭靖黄蓉在此书中退居配角，成为「社会规范」的化身——杨过小龙女与他们的冲突，本质是人性与社会的冲突

**变态人格的深度刻画**
- 李莫愁：爱情受挫→心狠手辣杀人如麻（情变型变态）
- 金轮法王、公孙止等反派各有特色

### 3.3 叙事技法
- **舞台集中法**：绝情谷成为小舞台，周伯通引入各色人物，众人各怀心思
- **暗笔**：王重阳通过周伯通之口和小龙女之口被勾勒——「匈奴未灭，何以家为」
- 引入心理学描写，白描与心理活动交织
- 武功个性化：「黯然销魂掌」是杨过性格的直接外化

### 3.4 语言风格
- 比《射雕》更多抒情笔调
- 「情」字贯穿全书——爱情、亲情（杨过对杨康的执念、对郭靖的复杂感情）、畸情（李莫愁）
- 文字技巧：可读作推理小说（金庸后记中提醒「读侦探小说，要理智地读」）
- 整体风格：庄中带谐、悲喜交织

### 3.5 特殊地位——金庸的转型里程碑
- 文学重心从「武侠小说」转向「小说」
- 人物从正邪分明转向正邪模糊
- 从「群像传记」转向「个体命运」
- 创作技法从「凌厉刚猛」走向「重剑无锋」

---

## 四、倚天屠龙记——三段体结构与人性的复杂迷宫

**定位**：叙事结构最庞大复杂的作品，对人性复杂与矛盾的深度开掘。

### 4.1 结构特征——三段体与断裂感

**三段式断裂结构**
- 第一段（第1-2回）：以张君宝（少年张三丰）和郭襄为主角，讲述郭襄寻杨过、觉远大师圆寂、张君宝开创武当派——实为《神雕》的余韵
- 第二段（第2-10回）：转向张翠山和殷素素——冰火岛遇谢逊、张无忌出生、武当山悲剧
- 第三段（第10-40回）：张无忌成为主角，武林争霸、明教崛起、抗元大业

**三段体的三大价值**
1. **释因**：暗线故事中的矛盾（谢逊与成昆的仇恨、六大派与明教的宿怨）是明线的前因
2. **造势**：通过大段前史铺垫，让读者对后续冲突产生期待
3. **蓄力**：张无忌的传奇色彩来自其身世背景的层层铺垫

*问题*：叙事结构明显割裂，主角变更三次，被部分学者批评为结构「混乱不连贯」

### 4.2 人物塑造——人性光谱最全的一部

**四层次人物体系**
- 主角张无忌：性格善良软弱、缺乏领袖魄力——金庸笔下最「被动」的主角（被命运推着走）
- 四大女主角：赵敏（智谋型）、周芷若（黑化型）、小昭（奉献型）、殷离（痴情型）——各代表爱情的一种面向
- 反派群像：成昆（阴谋家）、灭绝师太（极端正派→残忍）、朱长龄（为夺屠龙刀烧庄园设陷阱）
- 边缘高手：张三丰（一代宗师）、谢

Github ReposUpdated 1d agoRank 70

AionUi

Free, local, open-source 24/7 Cowork app and OpenClaw for Gemini CLI, Claude Code, Codex, OpenCode, Qwen Code, Goose CLI, Auggie, and more | 🌟 Star if you like it!

MCPOPENCLAW

Github ReposUpdated 6mo agoRank 70

activepieces

AI Agents & MCPs & AI Workflow Automation • (~400 MCP servers for AI agents) • AI Automation / AI Agent with MCPs • AI Workflows & AI Agents • MCPs for AI Agents

OPENCLAW

Github ReposUpdated 6mo agoRank 70

cherry-studio

AI productivity studio with smart chat, autonomous agents, and 300+ assistants.