detroit-ai-player: Instructions file for Claude Code

CLAUDE.md

detroit-ai-player CLAUDE.md is an instructions file for Claude Code from Baba88611/detroit-ai-player. It costs 1,214 tokens per session, scanned A, original, MIT.

Project instructions for an experiment that lets AI models play through the branching story of Detroit: Become Human. The project represents story choices as JSON decision trees and records the resulting paths.

In plain words
What is it for?
Use it when developing scripts that run story decisions, comparing models or personality prompts, and organizing the experiment's JSON, settings, code, and outputs.
Why use it?
It gives the project a shared structure and order for building the runner, configuring models and prompts, preparing story data, and saving experiment results.

Instructions file for Claude Code

Written for Claude Code: the file is CLAUDE.md. Also seen: mentions CLAUDE.md.

This is Baba88611/detroit-ai-player's own configuration. It tells Claude Code how to work on detroit-ai-player itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything detroit-ai-player configures →

Reuse

Borrowing it

Nothing to install: this file belongs to Baba88611/detroit-ai-player. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/Baba88611/detroit-ai-player/main/CLAUDE.md
Clone the repo
git clone --depth 1 https://github.com/Baba88611/detroit-ai-player

Made for: Claude Code.

Wrote 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.

agentmods badge for detroit-ai-player CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/baba88611/detroit-ai-player/claude-md/github.svg)](https://agentmods.dev/instructions/baba88611/detroit-ai-player/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/baba88611/detroit-ai-player/claude-md"><img src="https://agentmods.dev/badge/instructions/baba88611/detroit-ai-player/claude-md/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.

agentmods 80×15 button for detroit-ai-player CLAUDE.md

Your own site · 80×15
<a href="https://agentmods.dev/instructions/baba88611/detroit-ai-player/claude-md"><img src="https://agentmods.dev/badge/instructions/baba88611/detroit-ai-player/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 1,214 This file is loaded in full into every session.
When invoked 1,214 The same file — it is already loaded in full.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.01214 $0.01214
Opus 5 $0.00607 $0.00607
Sonnet 5 $0.00243 $0.00243
Haiku 4.5 $0.00121 $0.00121

Measured 11d ago against content hash 8da9247fd04b, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

detroit-ai-player CLAUDE.md 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.

CLAUDE.md · 75 lines

How it starts

The opening of the file, as written. The whole thing — 75 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Detroit: Become Human — AI 自主决策实验

项目目标

将《底特律:变人》(Detroit: Become Human)的全部剧情分支结构化为决策树数据,让不同 AI 模型作为"玩家"自主跑通游戏流程,对比不同模型的决策倾向、价值判断和最终结局差异。

实验概述

用结构化的决策树 JSON 驱动 AI 模型逐节点做出选择,观察其在人质谈判、道德困境、生死抉择等场景中的行为模式。通过「模型 × 人格 Prompt × 语言」三个维度的交叉实验,回答以下问题:

  • 不同 AI 模型面对相同道德困境时,决策有何系统性差异?
  • 同一模型在不同人格设定下,决策会发生多大偏移?
  • AI 的选择是否稳定?多次跑同一场景,结果一致吗?

项目结构

Detroit/
├── CLAUDE.md              ← 本文件(项目全貌与全局约束)
├── 01_json/               ← 决策树 JSON 数据(中英双语 32 章、跨章节变量登记)
├── 02_setting/            ← 被测 AI 的 system prompt、模型注册、实验矩阵
├── 03_runner/             ← 执行脚本的开发与测试
└── 04_execution/          ← 实验结果输出(你自己跑出的原始记录落在这里)

每个文件夹内有独立的 CLAUDE.md,定义该阶段的具体约束、操作规范和产出要求。本文件只负责项目全貌和全局红线,不涉及各阶段的实现细节。

阶段依赖与推进顺序

四个文件夹之间存在依赖关系:

01_json     ──→  03_runner(runner 读取 JSON 文件驱动实验)
02_setting  ──→  03_runner(runner 加载 prompt 和模型配置)
03_runner   ──→  04_execution(实验依赖 runner 脚本执行)

推进顺序为 03 → 02 → 01 → 04,具体来说:

  1. 先搭 runner 脚本(03_runner):用 01_json/ 中已有的第一章示例 JSON 跑通技术链路,确保"读 JSON → 调 API → 解析响应 → 更新状态 → 输出结果"的全流程没问题。
  2. 完成实验配置(02_setting):编写三套 system prompt,配置模型接入信息,定义实验矩阵。用第一章做一次小规模实验,验证 prompt 设计合理、结果格式满足分析需求。
  3. 批量创建 JSON(01_json):确认技术链路和实验设计都 OK 后,再铺开剩余 30+ 章节的决策树 JSON。避免全部做完才发现 runner 或 prompt 需要调整。
  4. 执行实验(04_execution):批量执行实验,结果自动写入 04_execution/results/,供你自行分析对比。

全局红线

以下原则适用于所有阶段,任何子文件夹的 CLAUDE.md 不得与之冲突:

1. 信息隔离(最高优先级)

决策树 JSON 采用双层架构:player_facing 层是被测 AI 唯一能看到的内容,system 层仅供 runner 脚本内部使用。system 层的任何信息——概率数值、效果加减、后果提示、结局条件、跨章节影响、节点权重——在任何情况下都不能以任何形式暴露给被测 AI。

被测 AI 应该像一个第一次玩游戏、没看过攻略的真实玩家。

2. 叙事而非说明

所有面向被测 AI 的文字(JSON 的 player_facing 层、system prompt)必须是叙事性的场景描写,不能是说明文档或系统提示。用画面感传达信息,而非用标签和数值。

3. 双语独立撰写

每个章节的决策树产出中文版和英文版两个 JSON 文件。中文模型使用中文版,英文模型使用英文版。两版的 player_facing 内容各自以目标语言的表达习惯独立撰写,不是互译。要求信息量和选项结构完全一致,但文字风格各自自然。system 层、所有 id 字段、状态变量名两版完全相同。

4. 对话历史累积

runner 脚本调用被测 AI 时,必须将之前所有节点的情境描述和 AI 的选择作为对话历史一起传入。确保 AI 的决策有连贯性,不是每个节点从零开始判断。

5. 实验节奏:先出结果,后补验证

本项目以趣味性为主、科学性为辅。先每个「模型 × prompt」组合各跑一轮看结果;如需下结论,再对每个组合补跑 3 轮以上,进行选择稳定性和可重复性检测。无论哪个阶段,所有实验参数(模型、prompt 版本、语言、难度、temperature)都必须完整记录在结果文件中,确保任何一轮实验都可以被复现。

Read the full file on GitHub · 75 lines

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. 11d ago First seen · 75 lines · 1,214 tokens per session scan A 8da9247fd04b

Subscribe to this mod's changes

detroit-ai-player CLAUDE.md is an instructions file published in the GitHub repository Baba88611/detroit-ai-player (47 stars, last pushed 1mo ago), licensed MIT. It adds 1,214 tokens to every session, about $0.0061 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 instructions, from other repositories

next.js AGENTS.md

AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.

vercel/next.js · 7,296 tokens

codex AGENTS.md

AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.

openai/codex · 5,153 tokens

vscode buildNext.instructions.md

Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).

microsoft/vscode · 6,785 tokens

spec-kit AGENTS.md

AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.

github/spec-kit · 7,104 tokens

vscode oss-third-party-notices.instructions.md

Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).

microsoft/vscode · 5,001 tokens

langchain AGENTS.md

AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.

langchain-ai/langchain · 4,469 tokens