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 agentmods add instructions/vikiboss/r2-web/claude-mdgit clone --depth 1 https://github.com/vikiboss/r2-webWrote 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/instructions/vikiboss/r2-web/claude-md)<a href="https://agentmods.dev/instructions/vikiboss/r2-web/claude-md"><img src="https://agentmods.dev/badge/instructions/vikiboss/r2-web/claude-md.svg" alt="Measured on agentmods" height="20"></a>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 | $0.01653 | $0.01653 |
| Opus 5 | $0.00826 | $0.00826 |
| Sonnet 5 | $0.00331 | $0.00331 |
| Haiku 4.5 | $0.00165 | $0.00165 |
Grade A, and why
r2-web 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 4d 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 — 138 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
本文档为 Claude Code (claude.ai/code) 提供仓库开发指南。
项目速览
R2 Web — 纯客户端 Cloudflare R2 存储桶文件管理器，零构建、零框架、零后端。
核心特性 文件上传、目录浏览、文件预览、文件操作、图片压缩、PWA、多语言（zh/zh_TW/en/ja）、浅色/深色主题、多选批量操作。
快速启动
npx serve src
# 或
python3 -m http.server 5500 --directory src
代码地图
快速定位表
| 任务 | 文件 |
|---|---|
| 修改文件名模板逻辑 | src/js/utils.js — applyFilenameTemplate |
| 修改图片压缩逻辑 | src/js/upload-manager.js — compressFile |
| 添加 i18n 文案 | src/js/i18n.js — const I18N |
| 修改按钮样式 | src/css/components.css — .btn |
| 添加设计 Token | src/css/tokens.css |
| 修改 R2 API 操作 | src/js/r2-client.js |
| 修改文件浏览逻辑 | src/js/file-explorer.js |
| 修改上传管理逻辑 | src/js/upload-manager.js |
| 多选批量操作逻辑 | src/js/file-explorer.js — toggleSelect / selectAll / clearSelection |
工具函数列表
以下函数定义在 src/js/utils.js，供各模块导入：
t(key, vars)— i18n 翻译（来自src/js/i18n.js）applyFilenameTemplate(tpl, file)— 文件名模板处理，占位符：[name]、[ext]、[hash:N]、[date:FORMAT]、[timestamp]、[uuid]、/compressFile(file, config, onStatus)— 图片压缩（定义在upload-manager.js）- PNG 特殊处理：直接优化缓冲区，不重新编码
- 自适应逻辑：压缩后更大则使用原文件
项目结构
r2-web/
├── readme.md — 项目说明、使用指南
├── package.json — 依赖声明（仅用于类型提示）
├── jsconfig.json — JSDoc 类型检查配置
└── src/ — 源码目录（即部署目录）
├── index.html — 应用外壳、import map、对话框模板
├── main.js — 入口（仅 new App()）
├── manifest.json — PWA 配置
├── style.css — 样式主入口（仅导入 css 子目录）
├── js/ — 业务逻辑模块
│ ├── app.js — 主协调器
│ ├── config-manager.js — 配置持久化、Base64 分享
│ ├── r2-client.js — S3 API 客户端
│ ├── ui-manager.js — 主题、Toast、对话框、Tooltip
│ ├── file-explorer.js — 目录导航、排序、分页、缩略图
│ ├── upload-manager.js — 上传、文件名模板、图片压缩
│ ├── file-preview.js — 图片/视频/音频/文本预览
│ ├── file-operations.js — 重命名、复制、移动、删除
│ ├── i18n.js — 多语言（zh/zh_TW/en/ja）
│ ├── constants.js — 常量
│ └── utils.js — 工具函数
└── css/ — 样式模块（CSS Layers）
├── reset.css — CSS Reset
├── tokens.css — 设计 Token（定义所有变量）
├── base.css — 全局基础样式
├── layout.css — 布局容器
├── components.css — 通用 UI 组件
├── utilities.css — 工具类
└── animations.css — 动画与过渡
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.
- 4d ago First seen · 138 lines · 1,653 tokens per session scan A b479a7f58e4f
r2-web CLAUDE.md is an instructions file published in the GitHub repository vikiboss/r2-web (455 stars, last pushed 2mo ago), licensed MIT. It adds 1,653 tokens to every session, about $0.0083 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.
Other instructions, from other repositories
rewind AGENTS.md
Instructions for pdugan20/rewind, covering rewind, work modes, development commands, agent configuration and architecture overview.
rewind CLAUDE.md
Instructions for pdugan20/rewind, a project described as: Personal data aggregation service. Syncs data from Strava, Last.fm, Discogs, Plex, Trakt, and Letterboxd into a centralized API.
agents AGENTS.md
AGENTS.md instructions for cloudflare/agents, covering agents.md, project overview, repository structure, nested agents.md files and setup.
joint CLAUDE.md
Claude Code instructions for clientIO/joint, covering claude.md, project overview, common commands, install dependencies and build distribution files (runs grunt).
cloudflare-workers-nextjs-saas-template AGENTS.md
AGENTS.md instructions for LubomirGeorgiev/cloudflare-workers-nextjs-saas-template, covering project context, vinext, general coding rules, control flow and comments.
claude-skills CLAUDE.md
Claude Code instructions for jezweb/claude-skills, covering claude skills, philosophy, directory structure, plugin anatomy (anthropic spec) and adding a new plugin.