Token导航 LogoToken导航

Agent Skill · 研究检索

vault-setup(保险库设置)

vault-setup 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

vault-setup 是研究检索类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为只读;许可证信息还不明确,第三方扫描已通过,按安装量在研究检索类的 73,675 个 Skill 里排第 66,424。

总安装 264周安装 11GitHub Stars 144下载 88核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
unknown

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:vault-setup(保险库设置)
来源仓库:https://github.com/earlyaidopters/second-brain
仓库路径:skills/vault-setup
安装命令:
npx skills add https://github.com/earlyaidopters/second-brain --skill vault-setup
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/earlyaidopters/second-brain --skill vault-setup

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

vault-setup 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中根据关键词、任务场景或来源线索快速定位候选结果。它可辅助从文档中提取内容、匹配字段或过滤无关信息,提升信息获取效率。

通过 npx skills add 命令从指定仓库安装,具体用法需结合 README 进一步确认。安装前建议检查权限范围和维护状态,避免触发不必要的联网或文件操作。适用宿主包括 Codex、Claude、Cursor、Gemini CLI,接入前应确认版本、权限和运行环境要求。

来源仓库内路径:skills/vault-setup

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 Codex、Claude、Cursor、Gemini CLI;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 npx skills add https://github.com/earlyaidopters/second-brain --skill vault-setu…;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它标注为只读,先用只读任务试一次,确认行为符合预期再扩大使用范围。

常见问题

6 条

vault-setup 怎么安装?

在终端执行 npx skills add https://github.com/earlyaidopters/second-brain --skill vault-setup。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

vault-setup 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

vault-setup 需要什么权限?

权限级别是只读。该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

vault-setup 是免费的吗?

公开资料没有标注明确许可证,收费和商用限制都无法确认。正式使用前请到来源仓库核对授权说明。

vault-setup 用的人多吗?

公开数据是累计安装 264 次,近一周安装 11 次,下载 88 次,GitHub 144 星。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

vault-setup 安全吗?

第三方扫描结果为通过,但这只覆盖已知的静态检查项,不能替代自己读一遍源码和安装命令。

值得选的地方和要先想清楚的地方

值得选的地方

  • 同时适配 4 种工具,换客户端不用重新找替代品。
  • 第三方扫描已通过,常见的高风险行为没有命中。
  • 标注为只读权限,试用阶段不容易误改本地文件。

先想清楚的地方

  • 许可证没有明确标注,能不能商用、能不能再分发都不确定。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 Codex,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。
  • 只想先试试效果、不希望它改动本地文件的人。

不太适合

  • 需要明确授权才能用到公司项目里的场景。
  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
vault-setup

保险库设置 · 本页

264144只读未明确通过

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

Vault Setup — Obsidian Configurator

Run from INSIDE the folder you want to become your Obsidian vault.

STEP 1 — One question, free text

Display this message exactly, then wait for their response:


Tell me about yourself in a few sentences so I can build your vault.

Answer these in whatever order feels natural:

  • What do you do for work?
  • What falls through the cracks most — what do you wish you tracked better?
  • Work only, or personal life too?
  • Do you have existing files to import? (PDFs, docs, slides)

No need to be formal. A few sentences is enough.


STEP 2 — Infer and preview, don't ask more questions

From their free-text answer, infer:

  • Their role (business owner / developer / consultant / creator / student)
  • Their primary pain point
  • Scope (work only / work + personal / full life OS)
  • Whether they have existing files

Then show a vault preview. Do NOT ask clarifying questions. Make smart inferences.

Here's your vault — ready to build when you are.

📁 [current directory name]
├── inbox/          Drop zone — everything new lands here first
├── daily/          Daily brain dumps and quick captures
├── [folder]/       [purpose based on their role]
├── [folder]/       [purpose based on their role]
├── [folder]/       [purpose based on their role]
├── projects/       Active work with status and next actions
└── archive/        Completed work — never deleted, just moved

Slash commands:
  /daily    — start your day with vault context
  /tldr     — save any session to the right folder
  /[role]   — [role-specific one-liner]

Type "build it" to create this, or tell me what to change.

Wait for confirmation before building anything.

STEP 3 — Build after confirmation

Once they say "build it", "yes", "go", "looks good", or similar:

Create folders

mkdir -p inbox daily [role folders] projects archive scripts \
  .claude/skills/daily .claude/skills/tldr .claude/skills/[role-command]

Role folder sets:

  • Business Owner → people/ operations/ decisions/
  • Developer → research/ clients/
  • Consultant → clients/ research/
  • Creator → content/ research/ clients/
  • Student → notes/ research/

If personal scope → also personal/

Open in Obsidian

open -a Obsidian "$(pwd)"

Write CLAUDE.md

Write directly to CLAUDE.md in the current directory:

# CLAUDE.md — [inferred role]'s Second Brain

## Who I Am
[2-3 sentences based on what they told you — specific, personal, written in first person as Claude describing its owner]

## My Vault Structure
[folder tree with one-line purpose per folder]

## How I Work
[3-4 bullet points inferred from their answers — capture style, main pain point, scope, what they want from AI]

## Context Rules
When I mention a decision → check [decisions or relevant folder] first
When I mention a person/client/project → look in [relevant folder]
When I ask you to write → read recent daily/ notes to match my voice
When something lands in inbox/ → ask if I want it sorted now

Write skill files

.claude/skills/daily/SKILL.md: Read today's daily note or create one. Check inbox/ for unprocessed files. Surface top 3 priorities. Ask: "What are we working on today?"

.claude/skills/tldr/SKILL.md: Summarize this conversation: decisions, things to remember, next actions. Save to the most relevant folder. Update memory.md.

Role-specific skill:

  • Business Owner → .claude/skills/standup/SKILL.md — briefing across projects, decisions, people
  • Developer → .claude/skills/project/SKILL.md — load a project's full context
  • Consultant → .claude/skills/client/SKILL.md — load a client's full context
  • Creator → .claude/skills/content/SKILL.md — read content folder, calibrate voice, develop idea
  • Student → .claude/skills/research/SKILL.md — pull all notes on a topic, synthesize

Write memory.md

# Memory

## Session Log
[Updated by Claude Code after each session]

## My Preferences
[Added as Claude learns them]

STEP 4 — Context injection question

After building, ask:

One last thing — how do you want your vault context loaded into Claude Code?

1. Global (recommended) — adds one line to ~/.claude/CLAUDE.md so your vault
   context loads automatically in every Claude Code session on this machine
2. Manual — I'll give you the line to paste into specific projects when you need it
3. Vault only — works automatically when you run claude from inside this folder

If global: Append to ~/.claude/CLAUDE.md (create if needed):

## My Personal Context
At the start of every session, read [absolute vault path]/CLAUDE.md for context about who I am, my work, and my conventions.

STEP 5 — Final output

Done. Your vault is live in Obsidian.

One manual step left:
  Obsidian → Settings → General → Enable Command Line Interface

Your slash commands:
  /daily    — run this tomorrow morning
  /tldr     — run this at the end of any session
  /[role]   — [one liner]

Have files to import?
  python scripts/process_docs_to_obsidian.py ~/your-files inbox/
  Then: "Sort everything in inbox/ into the right folders"

平台分布

Codex

31.87%
按下载量换算28

Claude

31.82%
按下载量换算28

Cursor

18.81%
按下载量换算17

Gemini CLI

8.99%
按下载量换算8

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills