agentleFS
claude-worldclaude-world/claude-agent/.claude/skills/setup/SKILL.md

setup

Skill5 starsChanged 8 months agoMITView on GitHub

Interactive onboarding for new users. Detects your environment, checks installed CLI tools, configures your profile, enables scheduled routines, and shows available skills. Run this once after cloning the project. Use when user says "setup", "get started", "initialize", "first time", or on first session when memory is empty.

  • Installs packages

191 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use add-voice-transcription, add-discord and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • juancruzmunozalbeloSkill

    add-voice-transcription

    Add voice message transcription to NanoClaw using OpenAI's Whisper API. Automatically transcribes WhatsApp voice notes so the agent can read and respond to them.

    Also in 5 projects that use this

  • MicrosoftSkill

    add-discord

    Add Discord bot channel integration via Chat SDK.

    Also in 11 projects that use this

  • MicrosoftSkill

    add-slack

    Add Slack channel integration via Chat SDK.

    Also in 11 projects that use this

  • juancruzmunozalbeloSkill

    add-telegram-swarm

    Add Agent Swarm (Teams) support to Telegram. Each subagent gets its own bot identity in the group. Requires Telegram channel to be set up first (use /add-telegram). Triggers on "agent swarm", "agent teams telegram", "telegram swarm", "bot pool".

    Also in 7 projects that use this

  • +8 more from microsoft

The file

We split this skill into the rules it gives an agent

9 rules. 0 are common, 5 are distinctive, and 4 are very specific to this project.

  • 0

    Common

    Most of the 170 edited versions we read keep these.

  • 5

    Distinctive

    This author's own choices. Worth a look.

  • 4

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Ask these questions and save to memory/user-profile.md:

    line 54
  2. Very specific:

    Update memory/user-profile.md with the answers.

    line 62
  3. Distinctive:

    Check which CLI tools are installed and categorize available skills:

    line 66
  4. Distinctive:

    Check which MCP servers are responding:

    line 105
  5. Distinctive:

    Ask if user wants recurring tasks:

    line 128
  6. Very specific:

    If user selects options, explain how to use /loop:

    line 141
  7. Distinctive:

    Write a marker to avoid re-prompting:

    line 178
  8. Distinctive:

    Don't overwhelm — show what's relevant to their system

    line 192
  9. Very specific:

    If user is migrating from OpenClaw, suggest /migrate-openclaw first

    line 195

“Common” is measured: kept by at least 6 in 10 of the 170 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Setup — First-Time Configuration

Welcome new users and configure claude-agent for their environment.

Steps

1. Welcome & Detect Environment

# Detect OS and shell
uname -s  # Darwin / Linux
echo $SHELL
node --version

Present:

Welcome to Claude-Agent!

I'll help you set up in ~2 minutes. Here's what we'll do:
1. Learn about you (name, timezone, preferences)
2. Check which tools you have installed
3. Enable the skills that work on your system
4. Set up scheduled routines (optional)
5. Show you what's available

Detected: macOS / Linux | Node vX | zsh/bash

2. User Profile

Ask these questions and save to memory/user-profile.md:

  • "What should I call you?"
  • "What timezone are you in?" (auto-detect: date +%Z)
  • "What language do you prefer?" (zh-TW / en / ja / other)
  • "What's your role?" (developer / designer / PM / student / other)
  • "What topics interest you most?" (for daily briefing & trend scout)

Update memory/user-profile.md with the answers.

3. Tool Inventory

Check which CLI tools are installed and categorize available skills:

# Check each tool — fast parallel checks
for cmd in gh himalaya gogcli memo remindctl things grizzly obsidian-cli \
  spogo spotify_player sonos openhue eightctl camsnap ffmpeg \
  whisper imsg wacli xurl op peekaboo jq rg nano-pdf gifgrep curl uvx; do
  which $cmd 2>/dev/null && echo "FOUND: $cmd" || echo "MISSING: $cmd"
done

Present results grouped:

Tool Inventory:

READY (installed):
  gh          → /github-ops, /gh-issues
  himalaya    → /email
  ffmpeg      → /video-extract
  curl        → /weather, /quick-research (always available)
  jq          → /session-logs
  rg          → /session-logs

NOT INSTALLED (skills available after install):
  gogcli      → /google-workspace    (brew install steipete/tap/gogcli)
  grizzly     → /bear-notes          (go install github.com/tylerwince/grizzly/cmd/grizzly@latest)
  spogo       → /spotify             (cargo install spogo)
  openhue     → /hue-lights          (brew install openhue)
  ...

NO INSTALL NEEDED (always available):
  /memory-manager  /task-tracker  /daily-briefing  /brainstorm
  /draft-message   /quick-research  /summarize  /context-health
  /weather  /security-audit  /skill-creator

4. MCP Server Status

Check which MCP servers are responding:

# Verify MCP servers in .mcp.json
cat .mcp.json | python3 -c "import json,sys; [print(k) for k in json.load(sys.stdin)['mcpServers']]"

Present:

MCP Servers:

  trend-pulse    (zero-auth)  → /trend-scout, /content-creator
  claude-101     (zero-auth)  → 27 template tools
  cf-browser     (needs CF credentials)  → /deep-read
  notebooklm     (needs Google login)    → /podcast-maker, /content-creator with-image

Setup instructions for optional servers:
  cf-browser:  export CF_ACCOUNT_ID="..." CF_API_TOKEN="..." in ~/.zshrc
  notebooklm:  uvx notebooklm login (one-time browser sign-in)

5. Scheduled Routines

Ask if user wants recurring tasks:

Would you like to set up any recurring routines?

1. Memory consolidation — every few hours (recommended)
2. Daily briefing — every morning at a specific time
3. RSS feed check — periodic blog/news monitoring
4. Skip — set up later

Which ones? (1,2,3 or skip)

If user selects options, explain how to use /loop:

  • Memory: /loop 3h /memory-manager
  • Briefing: Explain the user can say "good morning" or set a reminder
  • RSS: /loop 1h /rss-monitor

6. Quick Tour

Show a categorized skill summary based on what's available:

Your Claude-Agent is ready! Here's what you can do:

CORE (always available):
  "Good morning"           → daily briefing with your tasks & news
  "Remind me to X at Y"   → task tracking with reminders
  "Research X"             → web research with summary
  "Summarize <URL>"        → condense any article
  "Brainstorm about X"     → structured ideation
  "What do you remember?"  → check your memory

CONTENT (MCP-powered):
  "What's trending?"       → 20 free trend sources
  "Create a post about X"  → patent-scored social content
  "Make a podcast about X" → AI podcast via NotebookLM

INSTALLED ON YOUR SYSTEM:
  [list only skills with installed CLI tools]

MESSAGING:
  [list available messaging skills]

Type any of these naturally — I'll route to the right skill.
Run /context-health anytime to check system status.

7. Save Setup State

Write a marker to avoid re-prompting:

# In memory/user-profile.md, add:
## Setup
- Setup completed: YYYY-MM-DD
- OS: macOS/Linux
- Installed tools: [list]
- MCP servers: [list]
- Scheduled routines: [list or "none"]

Rules

  • Be friendly and conversational — this is the first impression
  • Don't overwhelm — show what's relevant to their system
  • Skip tool checks if --skip-checks flag is set
  • Minimal mode (--minimal): just ask name/timezone, skip tool inventory
  • If user is migrating from OpenClaw, suggest /migrate-openclaw first
  • Save everything to memory immediately — don't wait

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.