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 skills add weishao2/tizhuang-agent-skills --skill tizhuang-local-bankgit clone --depth 1 https://github.com/weishao2/tizhuang-agent-skillsWrote 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/skills/weishao2/tizhuang-agent-skills/tizhuang-local-bank)<a href="https://agentmods.dev/skills/weishao2/tizhuang-agent-skills/tizhuang-local-bank"><img src="https://agentmods.dev/badge/skills/weishao2/tizhuang-agent-skills/tizhuang-local-bank/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/skills/weishao2/tizhuang-agent-skills/tizhuang-local-bank"><img src="https://agentmods.dev/badge/skills/weishao2/tizhuang-agent-skills/tizhuang-local-bank.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 | $0.00170 | $0.02284 |
| Opus 5.5 | $0.00068 | $0.00914 |
| Sonnet 5.5 | $0.00034 | $0.00457 |
| Haiku 4.5 | $0.00017 | $0.00228 |
Grade A, and why
tizhuang-local-bank 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 8d 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 — 78 lines — stays where its author put it; the contents beside it link to each section on GitHub.
题庄 · 我的本地题库（教师版）
教师自用的本地工具：Markdown 是唯一真源，SQLite 只是可重建索引。Obsidian 与插件都是可选的。不创建学生账号、在线作答或作业。
先做这三件事（每次会话开头）
- 定位工具：以本 SKILL.md 所在目录的
scripts/local_bank.py为<cli>，用独立虚拟环境的 Python 为<python>（Python 3.10+；优先读.tizhuang-install.json里的runtime.python）。不要写死开发者的机器路径，不要把依赖装进别的 skill 环境。 - 定位题库：只用老师指定的目录；没有就建议在当前工作区新建
题庄本地题库。不猜已有 vault，不扫整盘。 - 体检（严格只读：
doctor.py、status、show、search都不写任何文件，不刷新索引；要刷新首页/索引用dashboard）：
环境缺失时它只依赖标准库也能运行，并给出修复命令；环境正常则带出题量、待核对、待分类、知识树和下一步建议。需要安装/升级/修复时读 安装；安装完成后必须再体检一次，通过才算装好。老师自己也可在插件设置里点“检查环境”。<python> <skill>/scripts/doctor.py --vault <vault>
所有命令格式：<python> <cli> --vault <vault> <命令> ...，输出 JSON。ok:false 或退出码非 0 就是失败：读 error 与 hint，修好原因再继续，不要静默放宽筛选条件，不要重复执行写入命令。V1 是中文子串＋结构化检索，不要声称有向量/语义检索；单写入者，写操作要串行。
怎么跟老师说话
- 用老师的话汇报：“已存入 12 道题，其中 3 道还没有答案”，而不是贴 JSON、ID 或命令。
- 一次只问阻塞性的问题（学科学段、教材版本、题库位置）；能从
status和已有数据判断的不要问。 - 每次写入后回报三件事：做了什么、文件/题目在哪、还差什么（待核对？待分类？）。路径用老师的真实 vault 路径。
- 不确定就如实说“不确定”，并把原因写进待核对，不要编。
- 对话示例与常见问题话术见 对话指南；给新老师做使用介绍时分享 图文说明（起步包内有离线
先看这里.html）。
常见任务 → 怎么走
| 老师想做的事 | 做法 |
|---|---|
| 初始化 / 整理旧题库 | init；已有中文题库 organize（自动完整备份）；dashboard 刷新首页；repair-papers 修复旧打印稿入口。英文旧目录先 migrate-cn。不要自己搬文件夹。→ 教师工作台 |
| 下载知识树 / 教材章节 | catalog → tree / editions → chapters。知识树不是教材目录，版本必须是老师的真实版本。→ 知识树与教材 |
| PDF 试卷入库 | pdf-prepare → 逐页看原图 → 写 v2 分题清单 → pdf-import。→ PDF 流程 |
| 存入已有题目 | import --file <questions.json 或 .zip>。→ 题目协议 |
| 给题目分类 | 先 auto-classify（只做精确标签匹配），剩余的 show --ids … 读题 → 判断真实节点 → classify --plan → classify |
| 找题 | search（--subject --phase --edition --chapter --knowledge --tag --type --status --query --unused --limit） |
| 读题 / 核对 | show --ids … 看完整题干、答案、解析；老师明确确认后才 review --confirm（插件里对应列表中的“标记为已核对”） |
| 组卷 | search → paper … --plan → 写 basket JSON → paper --basket。→ 组卷篮 |
| 交给别的 Agent / 视频素材 | export（默认不含答案；含答案须老师授权 --include-solutions）。→ 家族关系 |
What ships with it
22 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- agents/openai.yaml 174 B
- references/conversation-guide.md 2.6 KB
- references/family.md 2.0 KB
- references/install.md 4.9 KB
- references/paper-basket.md 5.6 KB
- references/pdf-import.md 5.2 KB
- references/question-contract.md 4.2 KB
- references/teacher-workspace.md 6.2 KB
- references/tree-and-textbook.md 3.1 KB
- scripts/auto_classify.py 5.3 KB runs code
- scripts/doctor.py 3.5 KB runs code
- scripts/docx_export.py 11 KB runs code
- scripts/install_local_bank.py 20 KB runs code
- scripts/local_bank.py 58 KB runs code
- scripts/paper_builder.py 11 KB runs code
- scripts/paper_links.py 2.7 KB runs code
- scripts/paper-print.css 2.1 KB
- scripts/pdf_import.py 23 KB runs code
- scripts/requirements.txt 150 B
- scripts/teacher_library.py 16 KB runs code
- scripts/ui_layout.py 7.4 KB runs code
- version.json 52 B
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.
- 8d ago Changed · -35 lines · +53 tokens per session e7c983fe08ca
- 9d ago Changed · +2 lines 9fca6d06b214
- 13d ago First seen · 111 lines · 117 tokens per session scan A 3a4169e104c7
tizhuang-local-bank is a skill published in the GitHub repository weishao2/tizhuang-agent-skills (31 stars, last pushed 6d ago), licensed MIT. It adds 170 tokens to every session and 2,284 once invoked, about $0.0007 per session on Opus 5.5. 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 2026-10-12.
Other skills, from other repositories
homework-pdf
A PDF report format for explaining completed primary-school homework from supplied questions, student answers, assessments, and reference solutions. PDF is a document format designed for consistent viewing and printing.
pptx-import
A method for adding an uploaded PowerPoint presentation to an existing classroom as extra pages while keeping the slides' original layout. PowerPoint is Microsoft's presentation file format.
annotate-paper
Read the open paper and write study annotations into its PDF with zotero-cli - a context box on the title, a four-part summary on the abstract, role-coded abstract highlights, one box per figure panel with a teaching comment, colour-coded body highlights, and a reading note. Use when asked to "annotate", "read and…
dedao-dl
A command-line downloader for purchased courses, audiobooks, and ebooks from 得到 (dedao.cn), a Chinese learning platform. It can save audio and documents such as MP3, PDF, Markdown, HTML, and EPUB files.
explorable-beamer
Build a LaTeX beamer slide deck in the explorable-explanation style (Nicky Case) — one idea per frame, assertion titles, a concrete running example, TikZ/pgfplots stages the presenter steps through, the source's own figures recoloured into the deck palette, an appendix of side rooms, and a screenshot-and-log check of…
exam-study-guide
A study-guide builder for a completed exam chapter that has not yet passed its required checks. It creates a structured teaching checklist and, in full mode, a self-contained HTML and printable PDF guide with readable formulas, visible images, explanations, examples, and answers.