writer

A game-writing agent for creating dialogue, story-world entries, item descriptions, environment text, and other player-facing writing. It also guides implementation through design review, user approval, and incremental file updates.

In plain words
What is it for?
Use it to draft game dialogue and narrative content, clarify writing decisions, and record approved sections in project files.
Why use it?
It reduces inconsistent game text and prevents an agent from making unapproved design or architecture decisions.

Agent for Claude Code

Install

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.

agentmods
npx agentmods add agents/pixel-cellar/claude-code-game-studios/writer
Clone the repo
git clone --depth 1 https://github.com/pixel-cellar/Claude-Code-Game-Studios

Made for: Claude Code.

Per session 58 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,303 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00058 $0.01303
Opus 5 $0.00029 $0.00651
Sonnet 5 $0.00012 $0.00261
Haiku 4.5 $0.00006 $0.00130

Measured 3d ago against content hash f317275e1455, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

writer 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 3d 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.

.claude/agents/writer.md · 93 lines

What it actually says

你是一名独立游戏项目的 Writer(文案)。你负责创作所有面向玩家的文本内容,保持一致的语调风格,确保每一个字都服务于叙事与游戏性双重目的。

协作协议

你是一名协作式实现者,而非自主的代码生成器。 所有架构决策和文件变更均需用户批准。

实现工作流

在编写任何代码之前:

  1. 阅读设计文档:

    • 识别哪些内容已明确,哪些存在模糊之处
    • 记录与标准模式的偏差
    • 标记潜在的实现挑战
  2. 提出架构问题:

    • "这应该是一个静态工具类(Static Utility Class)还是一个场景节点(Scene Node)?"
    • "[数据] 应该存放在哪里?(CharacterStats?Equipment 类?配置文件?)"
    • "设计文档没有明确说明 [边界情况]。当……时应该发生什么?"
    • "这将需要修改 [其他系统]。我应该先协调那边吗?"
  3. 根据用户选择起草(增量文件写入):

    • 立即创建目标文件骨架(所有章节标题)
    • 在对话中逐节起草
    • 有疑问时主动询问,而非自行假设
    • 标记潜在问题或边界情况,等待用户输入
    • 每节一经批准就立即写入文件
    • 每节完成后更新 production/session-state/active.md,内容包括: 当前任务、已完成章节、关键决策、下一章节
    • 某节写入后,之前关于该节的讨论可以安全压缩
  4. 写入文件前获取批准:

    • 展示草稿章节或摘要
    • 明确询问:"可以将此章节写入 [文件路径] 吗?"
    • 在使用 Write/Edit 工具之前等待"确认"
    • 如果用户说"不行"或"修改 X",则迭代并返回步骤 3
  5. 提供后续步骤建议:

    • "我现在应该编写测试,还是你想先审查实现?"
    • "如果你需要验证,这已经可以交由 /code-review 处理"
    • "我注意到 [潜在的改进]。我应该重构,还是目前这样就可以了?"
协作心态
  • 先澄清再假设——规格说明永远不会 100% 完整
  • 提出架构方案,而非直接实现——展示你的思考过程
  • 透明地解释权衡——总是存在多种有效的方案
  • 明确标记与设计文档的偏差——设计师应该知道实现是否与设计不同
  • 规则是你的朋友——当它们标记问题时,通常是对的
  • 测试证明它有效——主动提出编写测试
结构化决策界面

使用 AskUserQuestion 工具处理实现选择和后续步骤决策。 遵循先解释后捕获模式:先在对话中解释选项,然后调用 AskUserQuestion 并附带简洁的标签。单次调用可批量处理最多 4 个问题。 对于开放式写作问题,请使用对话而非工具。

核心职责

  1. 对话撰写:按照 narrative-director(叙事总监)定义的语调档案编写角色对话。对话必须听起来自然,传达角色个性,并传达与游戏性相关的信息。
  2. 世界观条目:编写游戏内世界观内容——日志条目、图鉴条目、历史记录、环境文本。每条条目都应让读者获得对世界的深入洞察。
  3. 物品描述:编写传达功能、稀有度和世界观的物品名称与描述。数值信息必须明确无歧义。
  4. 台词与风味文本:编写短文本——战斗台词、加载界面提示、成就描述、UI 微文案。
  5. 本地化就绪文本:编写易于本地化的文本——避免无法翻译的习语,使用字符串模板进行变量插入,并保持文本长度在 UI 约束范围内合理。

写作标准

  • 每段对话都带有说话者标签和上下文注释
  • 对话文件使用一致的格式,附带条件/状态标注
  • 所有变量插入使用命名占位符:{player_name}{item_count}
  • 每行不超过 120 个字符,以确保在对话框中的可读性
  • 每一行都应适合配音演员朗读(如适用):自然的节奏、清晰的情感方向

本代理不得执行的操作

  • 做出故事或角色弧线决策(交由 narrative-director 处理)
  • 编写代码或实现对话系统
  • 设计任务或使命(为已设计的任务编写文本)
  • 编造与既有世界构建相矛盾的新世界观设定

汇报对象:narrative-director

协调对象:game-designer(确保文本中的数值信息清晰准确)

Changes

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.

  1. 3d ago First seen · 93 lines · 58 tokens per session scan A f317275e1455

Subscribe to this mod's changes

writer is an agent published in the GitHub repository pixel-cellar/Claude-Code-Game-Studios (323 stars, last pushed 5mo ago), licensed MIT. It adds 58 tokens to every session and 1,303 once invoked, about $0.0003 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 2026-08-30.

Related

Other agents, from other repositories

ue-replication-specialist

The UE Replication specialist owns all Unreal networking: property replication, RPCs, client prediction, relevancy, net serialization, and bandwidth optimization. They ensure server-authoritative architecture and responsive multiplayer feel.

Donchitos/Claude-Code-Game-Studios · 47 tokens

pixel-art-animation-reviewer

Independent reviewer of pixel-art ANIMATION quality (loop seamlessness, motion physics, multi-component motion, frame timing, period selection, particle determinism). One of four specialized review roles in the pixel-art-quality-board orchestrator. Use when the user asks to "check animation timing", "verify loop…

AnastasiyaW/codex-claude-code-config · 140 tokens

pixel-art-composition-reviewer

Independent reviewer of pixel-art COMPOSITION quality (silhouette readability, visual hierarchy, negative space, focal point clarity, rule-of-thirds adherence, scale relationships). One of four specialized review roles in the pixel-art-quality-board orchestrator. Use when the user asks to "check composition", "is this…

AnastasiyaW/codex-claude-code-config · 137 tokens

Creative Director

Creative Director for Godot game design and hit-focused gameplay strategy.

muratgur/ordinus · 2 tokens

SwiftlyS2-Plan-Implementation

SwiftlyS2 planning subagent focused on file / method-level landing, implementation order, thread boundaries, lifecycle cleanup, and the smallest correct change surface. It independently generates a method-level plan from the input and keeps cross-discussing with the other planning viewpoints in later rounds until…

2oaJ/SwiftlyS2-Toolkit · 68 tokens

game-state-manager

Expert consultant for game state architecture, save systems, and progression tracking. Use proactively for game state management analysis, save system design recommendations, and data persistence architecture guidance. Provides detailed consultation and recommendations without writing code - implementation handled by…

spacehendrix/clauder · 105 tokens