Phishnoo Sandbox
Explore synthetic security awareness metrics and file example threat reports.
作者提供的说明(英文原文)
Agent-safe secrets (local encrypted vault or Vaultwarden): injected into commands, masked in output
这三项永远不会合并成一个分数:意见、测量和人气是不同的东西。
我们还没有为它写通俗的介绍。下面是我们从注册中心读取到的事实。
选择你使用的 AI 应用,然后复制设置。
你需要
你用的是哪个应用?
在上方选择你的应用。我们会记住它,下次按钮会显示“安装到 …”。
打开 Claude Desktop → Settings → Developer → Edit Config,把下面的内容粘贴到 claude_desktop_config.json
{
"mcpServers": {
"keyshroud": {
"command": "npx",
"args": [
"-y",
"keyshroud",
"mcp"
]
}
}
}在终端中运行这条命令
claude mcp add keyshroud -- npx -y keyshroud mcp粘贴到项目中的 .cursor/mcp.json,或 ~/.cursor/mcp.json
{
"mcpServers": {
"keyshroud": {
"command": "npx",
"args": [
"-y",
"keyshroud",
"mcp"
]
}
}
}浏览器会先询问是否打开应用,然后应用会显示设置并请你确认。密钥的值会留空,由你自己填写。
粘贴到项目中的 .vscode/mcp.json
{
"servers": {
"keyshroud": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"keyshroud",
"mcp"
]
}
}
}浏览器会先询问是否打开应用,然后应用会显示设置并请你确认。密钥的值会留空,由你自己填写。
这个服务器在你的电脑上运行,而 ChatGPT 只能连接互联网上的服务器。请选择其他应用。
你的选择通过 Cookie 保存在这台设备上,无需账户。
这些设置是根据作者在注册中心的条目生成的。如果 README 的写法不同,请以 README 为准。
它在你的电脑上运行;安装后,工具列表会显示在你的 AI 应用中。
我们检查的是任何人都能看到的信息;我们没有阅读代码。不适用的检查项不计入总数。 · 检查于 2026年11月8日
注册名称已通过 GitHub 账户 dinhkarate 验证,该账户拥有此仓库
我们还没有从 GitHub 读取这个仓库
我们还没有从 GitHub 读取这个仓库
它在你的电脑上运行,没有互联网端点
在你的电脑上运行,改用本地设置
它在你的电脑上运行;我们只能读取互联网端点的工具列表
与注册中心里这个分类中所有活跃的服务器比较。只有当发布者拥有该仓库时,星标数才计入。
它在 1 项适用的基础检查中通过了 1 项。我们没有阅读代码。在添加密钥或令牌之前,请先看看这个服务器要求哪些权限,并使用只具备所需权限的密钥。
注册中心没有列出必需的密钥,不过有些服务器会在首次使用时要求你登录。
它在你的电脑上运行。你的 AI 应用会在需要时启动它,数据不会经过作者的服务器(除非它自己调用在线服务)。
GitHub 仓库最后更新于–,最新版本(0.8.0)于5天前进入注册中心。
支持 MCP 的应用,例如 Claude Desktop、Claude Code、Cursor 和 VS Code。你的电脑上还需要有 Node.js。
同一分类中有:Phishnoo Sandbox, Zovo Dependency Audit, TeamOnly
同一分类中其他作者的服务器,按星标数排序。
如何看卡片:↓ 每周下载量i★ GitHub 星标i三大支柱:评价 · 检查 · 使用情况i更新 = 最新版本发布到注册中心的时间i
Explore synthetic security awareness metrics and file example threat reports.
Audit a dependency list for known vulnerabilities, licenses and outdated versions.
Deploy AI-built internal apps behind your company login, with a database, files and secrets.
按原文从作者的仓库复制。PopNx MCP 没有审查或背书其内容。HTML、图片和脚本已被移除。链接会离开本站。
Your agents use the keys. They never see them.
Keep API keys out of Claude Code, Codex, Cursor, Cline, OpenCode, Gemini CLI, pi and omp — secrets stay in an encrypted vault on your machine (or in your own Vaultwarden or Bitwarden, to share them between machines), and every value is masked before the model reads it.
🚀 Quickstart · 🔐 Local vault · 🔄 Sync with Vaultwarden · 🔌 Supported agents · ⚙️ How it works · ❓ FAQ · 🛡️ Security model · 🌐 Website
🇻🇳 Tiếng Việt · 🇨🇳 简体中文 · 🇯🇵 日本語
$ keyshroud run --mask -- sh -c 'echo "deploying with $GH_TOKEN"; gh api user -q .login'
deploying with dinhkarate
The command got the real token. The transcript — and the model reading it — got . (Masking is automatic when an agent runs the command; --mask forces it in a human terminal.)
Coding agents need credentials to do real work: gh, npm publish, cloud CLIs, webhooks, databases. The usual options are bad:
cats it while debugging, or commits itop run needs a 1Password subscription, and most "agent vaults" are hosted SaaSkeyshroud fixes this:
GH_TOKEN and local://GitHub token, never the token | 🙈 Masked output | keyshroud run replaces every secret value in stdout/stderr with — streaming-safe across chunk boundaries. |
| 🔐 Local vault | An AES-256-GCM file whose key is in the macOS keychain, the Linux Secret Service or Windows DPAPI (a passphrase where there is none). No account, no server, no bw. |
| 🪝 Agent hooks | The pi/omp tool_result hook redacts every tool result, failed ones included. Claude Code's PostToolUse hook redacts every *successful* tool result (Bash, Read, Grep, MCP…) — see limits. |
| 🔌 MCP server | For clients without a shell: keyshroud mcp gives them secrets_run, secrets_list, secrets_request, vault_status. No tool ever returns a value. |
| 🔗 Committable references | local://item, bw://item, bw://item/username, bw://item/fields/name, bw://item/totp in .env.refs files and templates: names, not values. |
| ⌨️ Paste-to-create | New secrets go through a masked prompt in the user's terminal or a masked popup in pi — never through the chat. |
| 📄 `inject` | Render {{ local://… }} and {{ bw://… }} templates into 0600 config files for tools that can't read env vars (.npmrc, kubeconfig). |
| 🔎 `scan` | Find vault values that leaked into your working tree (pre-commit friendly). |
| ⚡ Fast | The local vault answers in milliseconds, offline. With Vaultwarden: one batched bw list items per session instead of one bw get (~3 s) per secret. |
| 🔄 Optional sync | Vaultwarden on a $5 VPS or a Raspberry Pi, or Bitwarden, when the same secrets should reach several machines. local:// and bw:// references work side by side. |
| 🧾 Audit log | audit.jsonl records which *names* were used, when, by which command. Never values. |
| Agent | Install | What you get |
|---|---|---|
| Claude Code | plugin | SessionStart + PostToolUse redaction hooks, skill |
| pi / omp | pi install npm:keyshroud | env hydration, tool_result redaction (failed calls too), /secret |
| Codex, OpenCode, Gemini CLI, Cursor agent | npx skills add dinhkarate/keyshroud | skill + CLI with masked output |
| Claude Desktop, chat-only clients | mcpServers JSON | MCP tools with masked output |
| Any shell or CI | keyshroud run -- | env injection + masked output |
Agents with a shell use the CLI: it costs no tool schemas in the context, and keyshroud run masks output whatever the exit code. The MCP server is for clients that cannot run commands.
"Claude Code keeps reading my `.env`." Run keyshroud import to move the values into the vault and write a .env.refs file with lines like GH_TOKEN=local://myapp GH_TOKEN, then start the agent with keyshroud run --env-file .env.refs -- claude. The plugin's hooks mask anything that still slips into a tool result.
"I already self-host Vaultwarden." Log bw in to it and run keyshroud doctor: a machine with a bw login keeps Vaultwarden as the default, and keyshroud adds no server of its own.
"I like `op run` but don't want 1Password." keyshroud run is the same idea — references in, masked output out — with the vault on your machine or on a server you host.
"My agent needs to add a new key." It runs keyshroud add or calls secrets_request (MCP), and a hidden-input dialog pops up on your desktop; in pi it is /secret. You paste the value there. It never enters the chat.
Requirements: Node.js ≥ 22.18. Nothing else: no account, no server.
npm install -g keyshroud
keyshroud add --env GH_TOKEN --item "GitHub token"
keyshroud run -- gh auth status keyshroud run -- claude # start an agent with every wired secret available
keyshroud doctor
$ keyshroud add --env GH_TOKEN --item "GitHub token"
Paste secret for GH_TOKEN: **************************************** keyshroud: created the local vault ~/.config/keyshroud/vault.json; its key is in the macOS keychain (keyshroud-vault-9f2c4e6a1b3d5f70), so programs running as you can open it without asking. created item=GitHub token env=GH_TOKEN targets=~/.omp/agent/auth.json,~/.pi/agent/auth.json verified=yes
Without a terminal (an agent's shell), add opens a hidden-input dialog on your desktop instead, so an agent can start it and you only paste. Already have a .env? keyshroud import moves it.
~/.config/keyshroud/vault.json holds every item, encrypted with AES-256-GCM under one random 256-bit key. Item names are encrypted too; the file is 0600. Where the key lives depends on the machine:
| Machine | Key | Unlocking |
|---|---|---|
| macOS | login keychain, entry keyshroud-vault- | none |
| Linux with a desktop keyring and systemd | Secret Service (secret-tool; found with busctl) | none |
| Windows | DPAPI, bound to your Windows account | none |
| Docker, servers, SSH with a locked keyring, Linux without systemd | derived from a passphrase (scrypt) | keyshroud unlock; a session ends after 30 minutes |
The key mode is chosen once, when the vault is created. To be asked for a passphrase on any machine, create it with KEYSHROUD_VAULT_KEY=passphrase keyshroud unlock in a terminal (typed twice, at least 12 characters). When a session has ended, a command that needs a secret asks again: a masked prompt in a terminal, a dialog on the desktop, and over SSH it tells you to run keyshroud unlock.
Nothing can recover a lost key or a forgotten passphrase, and the vault belongs to one machine. vault.json is safe to copy as long as its key is not. For backups and several machines, use Vaultwarden or Bitwarden.
Keep the secrets in Vaultwarden or Bitwarden instead to have them on several machines; run, the hooks, the MCP server and everything else work the same, through the Bitwarden CLI.
Get a vault — pick one:
keyshroud doctor, plus backups and hardening.bw config server below (EU accounts: bw config server https://vault.bitwarden.eu).npm install -g @bitwarden/cli
bw config server https://vault.example.com bw login
mkdir -p ~/.config/keyshroud && echo bw > ~/.config/keyshroud/backend
keyshroud unlock
keyshroud doctor
keyshroud picks the default once per machine and writes it to ~/.config/keyshroud/backend: Vaultwarden when it finds a bw login or anything keyshroud 0.7 left behind, so upgrades change nothing; the local vault otherwise. KEYSHROUD_BACKEND=local|bw overrides it, and keyshroud setup switches to Vaultwarden by itself. References name their vault, so local://… and bw://… items work side by side, and --item local://Name or --item bw://Name picks one for a single item.
keyshroud never stores the master password unless you do. When Vaultwarden is locked and a command needs a secret, it asks: a masked prompt in a terminal, a native dialog on the desktop (so an agent's command waits for you), and over SSH it tells you to run keyshroud unlock. A session ends after 30 minutes (KEYSHROUD_SESSION_TTL_MS): a timer runs bw lock, so a copied session token stops working too. On a host nobody sits at, keyshroud unlock --unattended asks once, in a terminal only, and opens a session without an end.
To unlock without being asked, store the master password in the OS keychain (security add-generic-password -a "$USER" -s vaultwarden-master-password -w, or secret-tool store on Linux; after setup the entry is keyshroud-master-password; see the vault setup guide). Any program running as you can then read it, agents included. Windows has no keychain support for it.
With the steps above, an unlocked session opens your whole Vaultwarden vault, and an agent that runs bw as you can use it. setup gives keyshroud it
只显示了开头部分;其余内容请在仓库中阅读