Pinkbin is an open-source disk analyzer and cleaner that shows storage usage, lets AI explain unfamiliar folders, and supports controlled deletion through cleanup scripts. It is for Windows users who want to understand and reclaim disk space while keeping deletions scoped and sending them to the recycle bin by default. The catalogue commands, instruction, and setting support agent-assisted use of Pinkbin.
Borrowing it
Nothing to install: this file belongs to cccyd2003-qwq/pinkbin. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/cccyd2003-qwq/pinkbin/main/.claude/commands/add-scaffold.mdgit clone --depth 1 https://github.com/cccyd2003-qwq/pinkbinWrote 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/commands/cccyd2003-qwq/pinkbin/add-scaffold)<a href="https://agentmods.dev/commands/cccyd2003-qwq/pinkbin/add-scaffold"><img src="https://agentmods.dev/badge/commands/cccyd2003-qwq/pinkbin/add-scaffold/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.
<a href="https://agentmods.dev/commands/cccyd2003-qwq/pinkbin/add-scaffold"><img src="https://agentmods.dev/badge/commands/cccyd2003-qwq/pinkbin/add-scaffold.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00038 | $0.02054 |
| Opus 5 | $0.00019 | $0.01027 |
| Sonnet 5 | $0.00008 | $0.00411 |
| Haiku 4.5 | $0.00004 | $0.00205 |
Grade A, and why
add-scaffold 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 9d 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 — 171 lines — stays where its author put it; the contents beside it link to each section on GitHub.
为 app $ARGUMENTS 编写或重写一份 scaffold TOML。严格按以下 phase 顺序执行，每个 phase 完成后向用户报告进展并在必要时确认。绝不跳步——每一步都是上一次踩坑的产物（详见 CLAUDE.md）。
Phase 0 — Schema + UI 边界审查
读完这三个文件再开始任何实质工作：
crates/scaffold/src/lib.rs——Scaffold/Scope/Mode/Prompt定义、detect_for匹配逻辑、expand_env路径展开apps/desktop/src/components/Studio.tsx—— 当前 UI 如何渲染 scope（重点看 expanded card 部分调用了sc.scopes的什么字段）apps/desktop/src/types.ts—— TS 镜像类型
报告：当前 UI 是否已经能渲染你计划的 scope 形态？如果差距大，先去对齐 UI（或者明确告诉用户"TOML 写完了 UI 看不到"），再回来。
Phase 1-2 — 类别需求采集 → 落盘文档
$ARGUMENTS 属于哪个类别（messaging / browser / dev-tool / game / media / system / ai）？
- 如果
docs/scaffold-requirements/<category>.md已存在 → 读它，确认本应用是否已在范围内、L1/L2/L3 分级是否覆盖 - 如果不存在 → 复制
docs/scaffold-requirements/_TEMPLATE.md，用AskUserQuestion驱动结构化访谈，把答案落进去
访谈必问项：
- 类别下要覆盖哪些应用、优先级
- L1 缓存（可重生）默认行为：勾选/不勾选、保留 N 天 / 全量
- L2 历史（用户内容）默认保留期、是否暴露 Backup
- 多账号场景如何处理
- mode 默认值（recycle / quarantine / delete）
- 红线：除通用红线外还有哪些必须豁免
Phase 3 — 列出典型路径候选
为 $ARGUMENTS 给出所有合理候选：
- 安装路径（多版本、32/64 bit、用户自定义盘）
- 数据存储路径（默认 + 自定义盘）
- 缓存 / state 路径（
%APPDATA%/%LOCALAPPDATA%/~/.config/~/Library等） - 多平台差异（Windows / macOS / Linux 如适用）
报告：把候选清单贴给用户，告诉他下一步要在自己机器上确认。
Phase 4 — 用户确认实际路径
用 AskUserQuestion 让用户从候选里选 + 提供绝对路径。不要假设默认就够——很多人会改盘符。
Phase 5 — 实测勘测（Glob/ls 仅列文件夹名）
逐层下钻用户给的实际路径。严禁 Read 任何文件内容（隐私 + 不必要）。重点关注：
- 顶层布局
- 多账号目录命名（
wxid_*/<手机号>/accounts/<id>等） - 大头桶名称（image / video / file / cache / temp / msg / Backup / log / crashinfo / ...）
- "看起来是 DB 但其实不是" / "看起来不是 DB 但其实是" 的目录（
msg/在 WeChat 4.x 是用户文件，不是 DB）
把目录树贴给用户确认你的推断，再继续。
Phase 6 — 文件夹作用推断 + L1/L2/L3 映射
把 Phase 5 实测的每个目录映射到 L1（可重生缓存）/ L2（用户历史，可选清）/ L3（不可碰红线）。保守优先——拿不准的进 L3。
Phase 7 — 回环：修订需求文档
把 Phase 5-6 的发现追加到 docs/scaffold-requirements/<category>.md 的 "实测附录" 节。这一步在写 TOML 之前完成，不是之后。
如果实测推翻了 Phase 1-2 的假设（例：某个被列为红线的目录其实是用户内容），显式标注修订，并更新主 §2 红线节。
Phase 8 — 写 TOML
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.
- 9d ago First seen · 171 lines · 38 tokens per session scan A 4d9e35419f7e
add-scaffold is a command published in the GitHub repository cccyd2003-qwq/pinkbin (1,361 stars, last pushed 1mo ago), licensed MIT. It adds 38 tokens to every session and 2,054 once invoked, about $0.0002 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 commands, from other repositories
loci-sync
Sync information between this project and the Loci brain index. Keep full project memory in the project repo.
loci-scan
Refresh the current project's local Loci memory.
claude-tracker
List recent Claude Code sessions with live status.
gov-reuse
Discover reusable UK government code before building from scratch.
INDEX
18 custom slash commands in /.claude/commands/.
gpd:respond-to-referees
Structure a point-by-point response to referee reports for an explicit manuscript target or the current GPD manuscript.