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/jerry-terrasse/vimprove/claude-mdgit clone --depth 1 https://github.com/Jerry-Terrasse/vimproveWrote 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/jerry-terrasse/vimprove/claude-md)<a href="https://agentmods.dev/instructions/jerry-terrasse/vimprove/claude-md"><img src="https://agentmods.dev/badge/instructions/jerry-terrasse/vimprove/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 | $0.07046 | $0.07046 |
| Opus 5 | $0.03523 | $0.03523 |
| Sonnet 5 | $0.01409 | $0.01409 |
| Haiku 4.5 | $0.00705 | $0.00705 |
Grade A, and why
vimprove 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 5d 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 — 517 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
Project Overview
Vimprove 是一个交互式 Vim 学习网站。核心功能是通过浏览器中的"迷你 Vim 编辑器 + 关卡式练习"来教用户实际操作 Vim 命令。
当前状态: ✅ 重构完成。项目已从单文件原型(tmp/vimprove.html)重构为模块化的 React + TypeScript 架构。
课程范围: 已完成 Chapter 1-6(基础、进阶编辑、行内 find/till、文本对象、搜索/重构),共 28 节课。
版本管理: 版本号在 src/version.ts 和 package.json 中维护,CHANGELOG 见 README.md
Development Commands
npm run dev # Start dev server at http://localhost:3000
npm run build # Build for production
npm run preview # Preview production build
npm run lint # Run ESLint
npx vitest run --pool=threads # deprecated
# Note: please refer to section "Test Workflow" and use ./utils/vitest-quickcheck.sh to improve efficiency.
Test Workflow
- 默认并行:
npx vitest run --pool=threads(避免直接跑无过滤的npx vitest run,输出过长会淹没上下文) - 测试输出简化: 重要 使用
grep过滤最终结论,避免无用信息淹没上下文。如npx vitest run --pool=threads [-t <pattern>] 2>&1 | grep -EA20 "Failed Tests|Test Files" - 测试输出长度:
grep -EA20是使用的20行,所获信息不足时,可以根据实际情况适当增加,建议不超过50行 - Parity 单测输出节流:
npx vitest run --pool=threads -t "<pattern>" src/core/tests/exhaustiveTest.*.test.ts 2>&1 | grep -EA20 "Failed Tests|Test Files",避免海量 skip 日志淹没上下文 - exhaustive parity 分片:
src/core/tests/exhaustiveTest.{0..7}.test.ts使用getShardCases按索引取模分片(总 shard=8),并行示例:npx vitest run --pool=threads src/core/tests/exhaustiveTest.*.test.ts --maxWorkers=8 - 快速检查脚本:
bash utils/vitest-quickcheck.sh [<test_glob>](tap-flat + bail,默认跑全部,可传入路径/模式,成功输出 ok ✅,失败时列出前 5 条 not ok) - 深入排查(exhaustive parity):
- 生成 JSON 报告:
npx vitest run --pool=threads --reporter=json --outputFile tmp/vimParity-report.json src/core/tests/exhaustiveTest.*.test.ts - 查看摘要/聚合或按子串过滤:
python utils/vimParity-report-viewer.py tmp/vimParity-report.json ["keyword"...](keyword 为测试名片段,支持多个并且大小写不敏感;工具仅用于 ParityExhaustive)- 支持附加参数:
--feature paste-after-op、--limit 5、--sort name|feature|line、--details(输出完整断言)
- 支持附加参数:
- 反复调试单用例:
npx vitest run --pool=threads -t "<pattern>" src/core/tests/exhaustiveTest.*.test.ts 2>&1 | grep -EA20 "Failed Tests|Test Files"- 正则含特殊符号时建议单引号包裹并在内部转义,如
-t 'd\$P\.'
- 正则含特殊符号时建议单引号包裹并在内部转义,如
- 生成 JSON 报告:
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.
- 5d ago First seen · 517 lines · 7,046 tokens per session scan A ea55924b045c
vimprove CLAUDE.md is an instructions file published in the GitHub repository Jerry-Terrasse/vimprove (121 stars, last pushed 7mo ago), licensed MIT. It adds 7,046 tokens to every session, about $0.0352 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
codewithphp copilot-instructions.md
Instructions for dalehurley/codewithphp, covering copilot ai agent instructions for php-from-scratch, project overview, key conventions & patterns, developer workflows and integration points.
dotfiles CLAUDE.md
Instructions for joshukraine/dotfiles, covering claude.md, about this repository, repository structure, common commands and setup and installation.
real-world-rails CLAUDE.md
Claude Code instructions for steveclarke/real-world-rails, covering real world rails, how to use this repo and key details.
learn-faster-kit AGENTS.md
AGENTS.md instructions for hluaguo/learn-faster-kit, covering repository guidelines, project structure & module organization, build, test, and development commands, coding style & naming conventions and testing guidelines.
TapCanvas AGENTS.md
AGENTS.md instructions for anymouschina/TapCanvas, covering 适用范围, 项目目标与编码原则, 文件与模块化要求, 数据安全与高风险操作 and 思维与决策方法.
dotfiles CLAUDE.md
Instructions for IvanWng97/dotfiles, a project described as: macOS dotfiles managed with GNU stow — ghostty, neovim, zsh, tmux, starship, lazygit, homebrew.