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/veithly/vibeshell/claude-mdgit clone --depth 1 https://github.com/veithly/vibeshellWhat 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 | $0.01411 | $0.01411 |
| Opus 5 | $0.00705 | $0.00705 |
| Sonnet 5 | $0.00282 | $0.00282 |
| Haiku 4.5 | $0.00141 | $0.00141 |
Grade A, and why
vibeshell 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 2d 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 — 115 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md — VibeShell
What is this project?
VibeShell is a desktop SSH/SFTP terminal application. Think of it as a modern alternative to PuTTY/Tabby/Xshell, built with Tauri 2 (Rust) + React 18 (TypeScript).
Quick start
npx tauri dev # Dev mode (hot reload)
npm run build && cargo check # Verify both sides compile
unset CI CXX CC && npx tauri build --no-bundle # Release exe
Key files to know
| What | Where |
|---|---|
| App entry + layout | src/App.tsx |
| All Tauri commands registered | src-tauri/src/lib.rs (invoke_handler) |
| Backend command implementations | src-tauri/src/commands/*.rs |
| SSH client (russh) | src-tauri/src/ssh/client.rs |
| Session manager | src-tauri/src/session/manager.rs |
| Database schema + CRUD | src-tauri/src/storage/database.rs |
| Data models | src-tauri/src/storage/models.rs |
| Frontend Tauri wrapper | src/lib/tauri.ts (safeInvoke) |
| Zustand stores | src/stores/*.ts |
| Tailwind theme config | tailwind.config.js |
| Tauri config | src-tauri/tauri.conf.json |
Architecture in 30 seconds
Frontend (React + Zustand) → safeInvoke('command', args) → Tauri IPC → Rust command handler → business logic (SSH/SFTP/DB/tunnel)
Session I/O: mpsc::channel for input, broadcast::channel for output. Terminal uses xterm.js with WebGL renderer.
Database: SQLite via rusqlite. Tables: servers, groups, credentials, tunnel_configs, command_snippets, recordings, settings.
Gotchas & constraints
russh 0.44 (SSH library)
client::HandleandChannelare NOT Clone- For shared handle access:
Arc<tokio::sync::Mutex<Option<Handle>>> - For channel I/O:
channel.into_stream()thentokio::io::split() - Never try
.clone()on Handle or Channel — it won't compile
Theme system
- All colors use CSS custom properties via Tailwind:
bg-tokyo-bg,text-tokyo-fg,border-tokyo-bg-hl - Available theme colors:
tokyo-bg,tokyo-bg-dark,tokyo-bg-hl,tokyo-fg,tokyo-comment,tokyo-selection,tokyo-blue,tokyo-green,tokyo-red,tokyo-yellow,tokyo-magenta,tokyo-cyan - Do not use hardcoded hex colors like
#1a1b26— they break theme switching
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.
- 2d ago First seen · 115 lines · 1,411 tokens per session scan A 565ff95c9ace
vibeshell CLAUDE.md is an instructions file published in the GitHub repository veithly/vibeshell (76 stars, last pushed 6d ago), licensed MIT. It adds 1,411 tokens to every session, about $0.0071 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
oxideterm AGENTS.md
AGENTS.md instructions for AnalyseDeCircuit/oxideterm, covering oxideterm agent guidelines, 基本原则, 实现要求, 修改前检查 and 测试与验证.
ai-ops-mcp AGENTS.md
AGENTS.md instructions for Finn-Aigc/ai-ops-mcp, covering repository guidelines, project structure & module organization, build, test, and development commands, coding style & naming conventions and testing guidelines.
nextclaw AGENTS.md
Instructions for Peiiii/nextclaw, covering nextclaw ai 常驻内核, 产品愿景, 沟通与推进, 深思与目标模式 and 协作与 git 安全.
moonproxy-desktop AGENTS.md
Instructions for MoonProxyHQ/moonproxy-desktop, covering moonproxy, 功能, 技术栈, 目录结构 and 配置与数据存储.
CodePilot AGENTS.md
AGENTS.md instructions for op7418/CodePilot, covering agents.md, 项目协作定位, 共享规则入口（不在此重复，按需读）, 自检命令（codex 跑测试时） and codex review 规则.
MangoDisk AGENTS.md
AGENTS.md instructions for harry0703/MangoDisk, covering mangodisk contribution guidelines, product and architecture, implementation principles, naming and text and repository hygiene.