Skip to content

SkillMeat Onboarding Script

Estimated Duration: 30 minutes Audience: New SkillMeat users (individuals or small groups) Prerequisites: SkillMeat installed (skillmeat --version works) Setup: Participants on same or similar network

Pre-Training Checklist (5 minutes before)

  • [ ] Test your own SkillMeat installation
  • [ ] Have sample artifact specs ready (e.g., anthropics/skills/code-review)
  • [ ] Verify web interface works: skillmeat web dev
  • [ ] Gather participant emails for follow-up
  • [ ] Have written materials available (links or printouts)

Session Structure

Total Time: ~40 minutes (30 min + 10 min Q&A)

  1. Introduction (5 min)
  2. First Collection Walkthrough (10 min)
  3. Web Interface Demo (10 min)
  4. Marketplace Overview (5 min)
  5. Wrap-up and Resources (5 min)

Part 1: Introduction (5 minutes)

What to Say

"Welcome to SkillMeat! I'm [Your Name], and today I'm going to show you how to get up and running with SkillMeat in about 30 minutes.

By the end of this session, you'll be able to: - Create your first collection - Add artifacts from GitHub - Deploy them to your Claude projects - Browse the marketplace - Use the web interface

Whether you're working solo or with a team, SkillMeat helps you manage all your Claude artifacts - skills, commands, agents - in one place.

Let's start by making sure everyone has SkillMeat installed. Can everyone run this command in your terminal?"

Verification Step

Command for participants to run: 

skillmeat --version

What they should see: 

skillmeat, version 0.3.0-beta

If it fails: - "If you don't see the version, SkillMeat might not be installed. After this session, see the quick start guide for installation help." - "We'll come back to installation questions at the end."

Set Expectations

"We'll go hands-on, so feel free to follow along on your machine as we go. If you get stuck, just raise your hand and I'll help. Let's start!"

Part 2: First Collection Walkthrough (10 minutes)

Introduction

"First, we're going to create your first collection. A collection is like a library of all your Claude artifacts - think of it as your personal repository where you keep everything organized."

Step 1: Initialize Collection (3 minutes)

Show command: 

skillmeat init

What to explain:

"This command initializes a new collection. It creates a configuration file at ~/.skillmeat/collection.toml that stores information about all your artifacts.

Let me show you what happens..."

Run it and show output: 

Initialized collection at ~/.skillmeat/
Created collection.toml with default settings

Have participants try:

"Now you try - go ahead and run skillmeat init on your machine. You might get a message that it already exists, which is fine."

Wait for confirmations, then:

"Great! Everyone should see either 'Initialized' or 'Already exists'. Either way, you're good to go."

Step 2: Add First Artifact (4 minutes)

Explain what we're doing:

"Now we're going to add an artifact. Artifacts are the building blocks - they could be skills that add capabilities, commands that provide utilities, or agents that automate tasks.

Let me add a skill from GitHub. This skill is a code review assistant that analyzes code and provides feedback."

Show command: 

skillmeat add anthropics/skills/code-review

Explain the format:

"The format is username/repo/path[@version]. You can also add skills from local paths with ./path/to/skill."

Run it and show output: 

Added skill: code-review
Source: anthropics/skills/code-review
Version: latest
Scope: user
Status: Ready to deploy

Have participants try:

"Pick an artifact you're interested in. Want another code-related one? Try: 

skillmeat add anthropics/skills/code-review
Or search for something in the marketplace with: 
skillmeat marketplace-search productivity
Then add something that interests you."

Wait for participants to add, then:

"Excellent! You've just added your first artifact. Each artifact is versioned, so you can track updates and safely manage changes."

Step 3: List Artifacts (2 minutes)

Show command: 

skillmeat list

Example output: 

Collection: default
├── code-review (skill)
│   Source: anthropics/skills/code-review
│   Version: latest
│   Status: Ready
└── [any other artifacts added]

Explain:

"This shows you everything in your collection. The status tells you if each artifact is ready to deploy. Notice the version - it's tracking 'latest', which means when the upstream artifact updates, you can update with a simple command.

You can also see full details with skillmeat show <name>."

Part 3: Web Interface Demo (10 minutes)

Introduction

"Now let me show you the web interface. This is the beautiful dashboard where you can see all your artifacts, deploy them, and monitor everything happening."

Start Web Server (2 minutes)

Show command: 

skillmeat web dev

Explain while it starts:

"This starts a local web server. It should automatically open in your browser at http://localhost:3000."

Wait for it to load, then:

"You should see the dashboard now. If not, try opening http://localhost:3000 manually in your browser."

Demo: Collections Dashboard (3 minutes)

Show the dashboard:

"You're looking at the collections dashboard. This is your command center. On the left you can see all your collections and artifacts. On the right is a quick view showing statistics about your collection.

Walk through: 1. Collections list - "Click on an artifact to see its details" 2. Artifact card - Click on one and show the detail drawer 3. Metadata panel - "Here's all the information about this artifact" 4. Deploy button - "See that deploy button? We'll talk about that in a second"

Demo: Deploy Interface (3 minutes)

Show deployment UI:

"Let me click the deploy button to show you how easy deployment is."

Click Deploy:

"You would select which Claude Code project to deploy this to, confirm, and it's deployed. You'll see real-time progress with a progress bar and status updates.

After deployment, the artifact is available in your Claude projects - no copying files or manual installation needed."

Explain features: - "Real-time progress with Server-Sent Events" - "Shows what was deployed and when" - "You can deploy multiple artifacts at once" - "Easy rollback if you need to undo"

Demo: Analytics (2 minutes)

If time permits, scroll to analytics section:

"SkillMeat also tracks how you're using your artifacts. This helps you understand which artifacts are most useful, and when to clean up ones you're not using anymore.

The analytics are private - SkillMeat only tracks usage on your machine. No data leaves your computer."

Part 4: Marketplace Overview (5 minutes)

Introduction

"One of the cool features is the marketplace. You can browse artifacts created by others, install what's useful, and even publish your own."

Search the Marketplace

Show command: 

skillmeat marketplace-search productivity

Explain:

"This searches the public marketplace for artifacts related to 'productivity'. You'll see results with names, descriptions, and ratings from other users."

Show installing: 

skillmeat marketplace-install skillmeat-42

"Installing is super easy - one command. The artifact is downloaded, verified, and added to your collection."

Marketplace in Web UI

Switch to web UI:

"You can also browse the marketplace right here in the web interface. It's easier to explore visually than with command line - you can filter by category, see ratings and reviews, and install with one click."

Publishing Your Own

"Once you create your own artifacts, you can publish them to the marketplace so others can use them. We have a guide on publishing if you're interested."

Point to resource: - "See the docs at publishing guide for complete details"

Part 5: Wrap-Up and Resources (5 minutes)

Key Takeaways

"Let me recap what you've learned today:

  1. Create a collection - skillmeat init
  2. Add artifacts - skillmeat add <spec>
  3. See everything - skillmeat list
  4. Use the web interface - skillmeat web dev
  5. Deploy to Claude - Click deploy in web UI
  6. Browse marketplace - skillmeat marketplace-search

These are the core workflows. Everything else builds on these basics."

Resources for Learning More

"Here are the key resources for diving deeper:

Quick References: - CLI Commands Reference - Quick lookup for all commands - Web UI Shortcuts - Keyboard shortcuts and tips

Detailed Guides: - Web UI Guide - Full walkthrough of web interface - Team Sharing Guide - Share with teammates - Marketplace Guide - Publish and share

Getting Help: - Documentation: https://docs.skillmeat.dev - GitHub Discussions: https://github.com/miethe/skillmeat/discussions - Email: support@skillmeat.dev"

Next Steps

"Here's what I recommend you do next:

  1. Explore - Add a few more artifacts that look interesting
  2. Experiment - Try deploying one to a project
  3. Learn - Read the guides that match your needs
  4. Share - If you're on a team, check out the team sharing guide

And don't hesitate to reach out with questions!"

Collect Feedback

"Before we wrap up, could you answer a quick question? Take 30 seconds to fill out this feedback form: Training Feedback Form

Your feedback helps us improve training for everyone else."

Part 6: Q&A and Troubleshooting (10 minutes)

Q&A Format

"Now let's open it up for questions. What would you like to know?"

Common Questions and Answers

Q: What's the difference between skills, commands, and agents?

"Great question!

Skills add new capabilities to Claude - like code review, documentation, design.

Commands are utilities and tools - things like API clients, file formatters, testing frameworks.

Agents can autonomously perform tasks - like running workflows, automating processes, or making decisions.

All three are managed the same way in SkillMeat - you add them, deploy them, and track them."

Q: Where do the artifacts go when I deploy them?

"They go to your Claude Code projects. SkillMeat handles the deployment automatically. They're stored in the .claude/skills/ directory in your project."

Q: Can I use SkillMeat if I'm working solo?

"Absolutely! SkillMeat is designed for individuals and teams alike. Solo users benefit from version tracking, easy updates, and search. Teams benefit from that plus sharing and collaboration."

Q: What if I want to modify an artifact?

"You can! There are different update strategies: - Overwrite: Replace with upstream version - Merge: Keep your changes and merge new updates - Fork: Create your own variant

See the updating guide for details."

Q: How do I know when artifacts have updates available?

"Run skillmeat status to check. You can update with skillmeat update <name> when you're ready."

Installation Issues

If someone didn't get SkillMeat installed:

"I see you're having installation trouble. Here's what to do:

  1. Make sure you have Python 3.9+ installed: python --version
  2. Install with: pip install skillmeat
  3. Verify: skillmeat --version

The quick start guide has detailed step-by-step instructions for your OS.

Feel free to reach out to us after this for help: support@skillmeat.dev"

Web Interface Not Starting

"If the web interface won't start:

  1. Make sure Node.js 18+ is installed: node --version
  2. Try a different port: skillmeat web dev --web-port 3001
  3. Check if port 3000 is in use: lsof -i :3000

If you're still stuck, we have a troubleshooting guide."

Follow-Up Email Template

Send this after the training:

Subject: SkillMeat Onboarding - Next Steps & Resources

Hi [Participants],

Thanks for joining today's SkillMeat onboarding! Here are the resources we discussed plus some next steps.

Quick References: - CLI Commands: ./cli-cheat-sheet.md - Web UI Shortcuts: ./web-ui-shortcuts.md - Common Workflows: ./common-workflows.md

Detailed Guides (read these for deeper learning): - Web UI Guide: https://github.com/miethe/skillmeat/docs/user/guides/web-ui-guide.md - Team Sharing Guide: https://github.com/miethe/skillmeat/docs/user/guides/team-sharing-guide.md - Marketplace Guide: https://github.com/miethe/skillmeat/docs/user/guides/marketplace-usage-guide.md - MCP Management: https://github.com/miethe/skillmeat/docs/user/guides/mcp-management.md

Recommended Next Steps: 1. Add 2-3 artifacts that interest you 2. Deploy one to a test project 3. Explore the web interface more 4. Read one detailed guide that matches your needs

Support: - Full Documentation: https://docs.skillmeat.dev - GitHub Discussions: https://github.com/miethe/skillmeat/discussions - Email: support@skillmeat.dev - Office Hours: Thursdays 2-3pm PT

Your Feedback: Please take 1 minute to complete our training feedback form. Your input helps us improve!

Questions? Reply to this email or ask in GitHub Discussions.

Best, SkillMeat Team

Facilitation Tips

Keep It Interactive

  • Have participants try commands as you demonstrate
  • Encourage questions throughout
  • Ask "what do you see?" to check understanding
  • Pair experienced users with newcomers

Adjust Pace

  • If group is struggling: Slow down, skip web demo, focus on CLI
  • If group is advanced: Expand on MCP and marketplace
  • Have "stretch topics" ready for fast movers

Handle Distractions

  • "Great question, let me note that and we'll cover it in the resources section"
  • Keep time sacred - don't let Q&A run over
  • Schedule office hours for extended discussions

Engagement Strategies

  • Start with hands-on (not just demo)
  • Show a real use case they care about
  • Use humor and relatable examples
  • Celebrate wins ("Nice, you added your first artifact!")

Recovery Techniques

  • If demo breaks: "Let me show you the command line - it's faster anyway"
  • If someone is stuck: Have a buddy help while you continue
  • If lost focus: "Let's recap - here's what we've done so far..."

Success Metrics

After this 30-minute session, participants should be able to:

  • [ ] Run skillmeat --version (verification)
  • [ ] Initialize a collection: skillmeat init
  • [ ] Add an artifact: skillmeat add <spec>
  • [ ] List artifacts: skillmeat list
  • [ ] Start web UI: skillmeat web dev
  • [ ] Navigate web dashboard
  • [ ] Understand artifact types
  • [ ] Know where to find help

Variations by Audience

For Teams

  • Emphasize team sharing section
  • Show vault setup examples
  • Discuss governance and policies
  • Highlight merge/conflict resolution

For Developers

  • Expand marketplace publishing section
  • Show artifact structure
  • Discuss versioning strategies
  • Cover security best practices

For Admins

  • Skip marketplace, focus on MCP
  • Show observability overview
  • Discuss deployment at scale
  • Cover troubleshooting tools

Version: 1.0 Last Updated: November 17, 2025 Questions? See Trainer Guide