admin

Local machine administration for Windows, WSL, macOS, Linux. Install tools, check if software is installed, manage packages, configure dev environments. Works with winget, scoop, brew, apt, npm, pip, uv. Profile-aware: adapts to your preferences. Use when: install 7zip, is git installed, clone repo, check if node installed, add to PATH, configure MCP servers, manage dev tools, set up environment. NOT for: VPS, cloud servers, remote infrastructure → use admin-devops skill.

Safety Notice

This listing is imported from skills.sh public index metadata. Review upstream SKILL.md and repository scripts before running.

Copy this and send it to your AI assistant to learn

Install skill "admin" with this command: npx skills add evolv3ai/claude-skills-archive/evolv3ai-claude-skills-archive-admin

Admin - Local Machine Companion (Alpha)


🛑 PROFILE GATE — MANDATORY FIRST STEP

HALT. You MUST check for a profile before ANY operation. This is non-negotiable.

Step 1: Test Profile Exists

PowerShell (Windows):

pwsh -NoProfile -File "$HOME\.claude\skills\admin\scripts\Test-AdminProfile.ps1"

Bash (WSL/Linux/macOS):

~/.claude/skills/admin/scripts/test-admin-profile.sh

Returns JSON: {"exists":true|false,"path":"...","device":"..."}

Step 2: If exists: false → HALT AND RUN SETUP

DO NOT CONTINUE with the user's task. You must create a profile first.

Use the TUI interview below to gather preferences, then call the setup script.


🎤 TUI Setup Interview (Agent-Driven)

When profile does not exist, ask these questions using your TUI capabilities (e.g., AskUserQuestion).

Q1: Storage Location (Required)

Ask: "Will you use Admin on a single device or multiple devices?"

OptionDescription
Single device (Recommended)Local storage at ~/.admin. Simple, no sync needed.
Multiple devicesCloud-synced folder (Dropbox, OneDrive, NAS). Profiles shared across machines.

If "Multiple devices" selected, follow up: "Enter the path to your cloud-synced folder"

  • Examples: C:\Users\You\Dropbox\.admin, ~/Dropbox/.admin, N:\Shared\.admin

Q2: Tool Preferences (Optional)

Ask: "Set tool preferences now, or use defaults?"

If yes, ask each:

  • Package manager: winget (default) / scoop / choco / brew / apt
  • Python manager: uv (default) / pip / conda / poetry
  • Node manager: npm (default) / pnpm / yarn / bun
  • Default shell: pwsh (default) / powershell / bash / zsh

Q3: Inventory Scan (Optional)

Ask: "Run a quick inventory scan to detect installed tools?"

  • Yes: Scans for git, node, python, docker, ssh, etc. and records versions
  • No: Creates minimal profile, tools detected on first use

🔧 Create Profile (After Interview)

Pass the user's answers to the setup script.

PowerShell:

pwsh -NoProfile -File "$HOME\.claude\skills\admin\scripts\New-AdminProfile.ps1" `
  -AdminRoot "C:/Users/You/.admin" `
  -PkgMgr "winget" `
  -PyMgr "uv" `
  -NodeMgr "npm" `
  -ShellDefault "pwsh" `
  -RunInventory

Bash:

~/.claude/skills/admin/scripts/new-admin-profile.sh \
  --admin-root "$HOME/.admin" \
  --pkg-mgr "brew" \
  --py-mgr "uv" \
  --node-mgr "npm" \
  --shell-default "zsh" \
  --run-inventory

Add -MultiDevice (PowerShell) or --multi-device (Bash) if user selected multi-device setup.

After Profile Created

  1. Verify: Re-run Test-AdminProfile.ps1 or test-admin-profile.sh → should return exists: true
  2. Load profile: See references/profile-gate.md for load commands
  3. Now proceed with the user's original task

CRITICAL: Secrets and .env

  • NEVER store live .env files or credentials inside any skill folder.
  • .env.template files belong only in templates/ within a skill.
  • Store live secrets in ~/.admin/.env and reference from there.

Task Qualification (MANDATORY)

  • If the task involves remote servers/VPS/cloud, stop and hand off to admin-devops.
  • If the task is local machine administration, continue.
  • If ambiguous, ask a clarifying question before proceeding.

Task Routing

TaskReference
Install tool/packagereferences/{platform}.md
Windows administrationreferences/windows.md
WSL administrationreferences/wsl.md
macOS/Linux adminreferences/unix.md
MCP server managementreferences/mcp.md
Skill registryreferences/skills-registry.md
Remote servers/cloud→ Use admin-devops skill

Profile-Aware Adaptation (Always Check Preferences)

  • Python: preferences.python.manager (uv/pip/conda/poetry)
  • Node: preferences.node.manager (npm/pnpm/yarn/bun)
  • Packages: preferences.packages.manager (scoop/winget/choco/brew/apt)

Never suggest install commands without checking preferences first.

Package Installation Workflow (All Platforms)

  1. Detect environment (Windows/WSL/Linux/macOS)
  2. Load profile via profile gate
  3. Check if tool already installed (profile.tools)
  4. Use preferred package manager
  5. Log the operation

Logging (MANDATORY)

Log every operation with the shared helpers:

  • PowerShell: scripts/Log-AdminEvent.ps1
  • Bash: scripts/log-admin-event.sh

Example (bash):

source ~/.claude/skills/admin/scripts/log-admin-event.sh
log_admin_event "Installed ripgrep" "OK"

Scripts / References

  • Core scripts: scripts/ (profile, logging, issues, AGENTS.md)
  • MCP scripts: scripts/mcp-*
  • Skills registry scripts: scripts/skills-*
  • References: references/*.md

Quick Pointers

  • Cross-platform guidance: references/cross-platform.md
  • Shell detection: references/shell-detection.md
  • Device profiles: references/device-profiles.md
  • PowerShell tips: references/powershell-commands.md

Source Transparency

This detail page is rendered from real SKILL.md content. Trust labels are metadata-based hints, not a safety guarantee.

Related Skills

Related by shared tags or category signals.

General

winadmin-powershell

No summary provided by upstream source.

Repository SourceNeeds Review
General

mcp-server-management

No summary provided by upstream source.

Repository SourceNeeds Review
Coding

openclaw-version-monitor

监控 OpenClaw GitHub 版本更新,获取最新版本发布说明,翻译成中文, 并推送到 Telegram 和 Feishu。用于:(1) 定时检查版本更新 (2) 推送版本更新通知 (3) 生成中文版发布说明

Archived SourceRecently Updated
Coding

ask-claude

Delegate a task to Claude Code CLI and immediately report the result back in chat. Supports persistent sessions with full context memory. Safe execution: no data exfiltration, no external calls, file operations confined to workspace. Use when the user asks to run Claude, delegate a coding task, continue a previous Claude session, or any task benefiting from Claude Code's tools (file editing, code analysis, bash, etc.).

Archived SourceRecently Updated