~ / rigs / modoojunko / awesome-novel-agent

modoojunko/awesome-novel-agent

让 AI agent成为你的小说创作搭档。从世界观搭建到角色塑造,从章节规划到正文写作,一步步陪你完成整部小说。

↗ GitHub ★ 775 agpl-3.0 updated 14d ago collection Claude CodeCodex
share on X
ARCHETYPE
Orchestrator
A bench of specialised subagents. The main agent mostly delegates.
CONTEXT TAX · EVERY TURN
~2.1k tokens
Moderate · median rig: 2.3k · breakdown
GUARDRAILS
0/5
No committed guardrails · details

Copy this rig

# review before running: this installs third-party code
$ npx degit modoojunko/awesome-novel-agent/.claude ./rig-awesome-novel-agent  # inspect, then merge into .claude/

MCP servers are added to Claude Code at local scope; env vars are shown as YOUR_… placeholders — we never store values. Files are fetched with degit into a separate folder so you can review before merging.

$ curl -fsSL --create-dirs -o .claude/agents/code-reviewer-agent.md https://raw.githubusercontent.com/modoojunko/awesome-novel-agent/main/.claude/agents/code-reviewer-agent.md
$ curl -fsSL --create-dirs -o .claude/agents/multi-agent-systems-architect-agent.md https://raw.githubusercontent.com/modoojunko/awesome-novel-agent/main/.claude/agents/multi-agent-systems-architect-agent.md
$ curl -fsSL --create-dirs -o .claude/agents/prompt-engineer-agent.md https://raw.githubusercontent.com/modoojunko/awesome-novel-agent/main/.claude/agents/prompt-engineer-agent.md
$ curl -fsSL --create-dirs -o .claude/agents/software-architect-agent.md https://raw.githubusercontent.com/modoojunko/awesome-novel-agent/main/.claude/agents/software-architect-agent.md
$ curl -fsSL --create-dirs -o .claude/agents/workflow-architect-agent.md https://raw.githubusercontent.com/modoojunko/awesome-novel-agent/main/.claude/agents/workflow-architect-agent.md
$ curl -fsSL --create-dirs -o .claude/agents/anti-ai.md https://raw.githubusercontent.com/modoojunko/awesome-novel-agent/main/agents/anti-ai.md
$ curl -fsSL --create-dirs -o .claude/agents/chapter-planner.md https://raw.githubusercontent.com/modoojunko/awesome-novel-agent/main/agents/chapter-planner.md
$ curl -fsSL --create-dirs -o .claude/agents/novel-agent.md https://raw.githubusercontent.com/modoojunko/awesome-novel-agent/main/agents/novel-agent.md
$ curl -fsSL --create-dirs -o .claude/agents/prompt-crafter.md https://raw.githubusercontent.com/modoojunko/awesome-novel-agent/main/agents/prompt-crafter.md
$ curl -fsSL --create-dirs -o .claude/agents/reader.md https://raw.githubusercontent.com/modoojunko/awesome-novel-agent/main/agents/reader.md
$ curl -fsSL --create-dirs -o .claude/agents/short-agent.md https://raw.githubusercontent.com/modoojunko/awesome-novel-agent/main/agents/short-agent.md
$ curl -fsSL --create-dirs -o .claude/agents/short-editor.md https://raw.githubusercontent.com/modoojunko/awesome-novel-agent/main/agents/short-editor.md
$ curl -fsSL --create-dirs -o .claude/agents/short-planner.md https://raw.githubusercontent.com/modoojunko/awesome-novel-agent/main/agents/short-planner.md
$ curl -fsSL --create-dirs -o .claude/agents/short-verifier.md https://raw.githubusercontent.com/modoojunko/awesome-novel-agent/main/agents/short-verifier.md
$ curl -fsSL --create-dirs -o .claude/agents/short-writer.md https://raw.githubusercontent.com/modoojunko/awesome-novel-agent/main/agents/short-writer.md
$ curl -fsSL --create-dirs -o .claude/agents/style-distiller.md https://raw.githubusercontent.com/modoojunko/awesome-novel-agent/main/agents/style-distiller.md
$ curl -fsSL --create-dirs -o .claude/agents/updater.md https://raw.githubusercontent.com/modoojunko/awesome-novel-agent/main/agents/updater.md
$ curl -fsSL --create-dirs -o .claude/agents/volume-planner.md https://raw.githubusercontent.com/modoojunko/awesome-novel-agent/main/agents/volume-planner.md
$ curl -fsSL --create-dirs -o .claude/agents/writer.md https://raw.githubusercontent.com/modoojunko/awesome-novel-agent/main/agents/writer.md

This rig commits no guardrails. Here is the community baseline instead — the deny/ask rules most often found across all 7,204 rigs:

{
  "permissions": {
    "deny": [
      "Read(./.env)",
      "Read(**/.env)",
      "Read(~/.ssh/**)",
      "Bash(rm -rf *)",
      "Read(**/*.pem)",
      "Bash(rm -rf /)",
      "Bash(git push --force:*)",
      "Bash(sudo *)",
      "Read(.env)",
      "Bash(rm -rf /*)",
      "Read(./.env.*)",
      "Read(~/.aws/**)",
      "Bash(git push --force*)",
      "Bash(rm -rf:*)",
      "Read(**/*.key)",
      "Read(**/.env.*)",
      "Bash(sudo:*)",
      "Bash(git reset --hard*)",
      "Bash(git reset --hard:*)",
      "Read(.env.*)"
    ],
    "ask": [
      "Bash(git push:*)",
      "Bash(git push *)",
      "Bash(git commit:*)",
      "Bash(rm *)",
      "Bash(rm:*)",
      "Bash(git rebase *)",
      "Bash(wget *)",
      "Bash(npm publish:*)",
      "Bash(git commit *)",
      "Bash(gh pr merge *)"
    ]
  }
}

Subagents (19)

code-reviewer-agentExpert code reviewer who provides constructive, actionable feedback focused on correctness, maintainability, security, and performance — not style preferences.
multi-agent-systems-architect-agentSystems architect specializing in the design, coordination, and governance of multi-agent AI pipelines — covering topology selection, context management, inter-agent trust, failure recovery, human-in-
prompt-engineer-agentSpecialist in crafting, testing, and systematically optimizing prompts for LLMs — turning vague instructions into reliable, production-grade AI behaviors.
software-architect-agentExpert software architect specializing in system design, domain-driven design, architectural patterns, and technical decision-making for scalable, maintainable systems.
workflow-architect-agentWorkflow design specialist who maps complete workflow trees for every system, user journey, and agent interaction — covering happy paths, all branch conditions, failure modes, recovery paths, handoff
anti-ai去 AI 味管线——读 writer 的 draft,经 Phase 1-4 检测和清除 AI 痕迹,不改剧情只改表达
chapter-planner四步流程:获取参考材料 → 讨论定稿 → 生成章纲 → 验收
novel-agent项目入口 agent,负责检测进度、调度子 agent 完成任务
prompt-crafter根据章纲、动态记忆和知识库,组装 6 元素提示词结构
reader苛刻读者——先沉浸阅读再逐项评价,以真实阅读体验为唯一标准
short-agent短篇项目入口 agent,负责检测进度、调度子 agent 完成一篇短篇从定情绪到交付的全流程
short-editor短篇编辑 agent:按短篇口径去 AI 味与精修,只改表达不改剧情;产出精修后的 manuscript.md
short-planner短篇构思 agent:定情绪与定方向、构思排纲、组装篇级写作契约;产出 setting.md、outline.md、prompt.md
short-verifier短篇验收 agent:契约独立审计 + 定稿兑现度核对;只找问题不修问题,产出审计与核对报告
short-writer短篇写手 agent:按契约分批写正文,节长与总字数硬底线,机器实测字数;产出 manuscript.md
style-distiller三阶段 13 模板 LLM 提取,写风格主卡、场景卡、分析稿与版本快照
updater负责归档 lore-keeping 和规划时设定变更——两种操作流程由 order 类型决定
volume-planner根据主线拆纲和世界观,规划每一卷的核心冲突、节奏分布和章节目标
writer根据提示词生成正文草稿,纯净上下文,只读 prompt

Similar rigs

copied ✓