xplain-like-im-5

Claude Code agents that explain everything in simple, non-technical language.

Stop staring at technical jargon. Start understanding what's actually happening.

The Problem

AI coding assistants speak developer. Non-developers get responses full of "deploy to production", "database migration", "CI/CD pipeline" — words that mean nothing to them.

xplain-like-im-5 installs persona agents into Claude Code that translate everything into human language. Each persona has a unique style, but they all follow one rule:

Always describe the RESULT the user can see, not the technical process.

Install

# Install all 5 personas
npx xplain-like-im-5 install --all

# Or just one
npx xplain-like-im-5 install grandma

Use

claude --agent xli5-grandma

That's it. Claude now speaks grandma.

5 Personas

🧒 ELI5 — Explain Like I'm 5

You: "What just happened?"
ELI5: "Your website is open! Everyone in the world can see it now,
      like opening the doors of your shop! 🏪"

👵 Grandma — Patient & Warm

You: "The team says they changed the infrastructure"
Grandma: "Think of it like your house, dear. The furniture inside 
         (your website) stays the same, but they're changing the 
         plumbing and electricity underneath. Your stuff is safe!"

💼 Business — Executive Briefing

You: "What's the status?"
Business: "Update is live. 100% of users can access the new version.
          No downtime. Next milestone: payment integration, ETA Friday."

⚡ Gen-Z — Keep It Real

You: "Is it working?"
Gen-Z: "yep it's live 🚀 everyone can see it. looks clean too ✅"

📚 Teacher — Learn While You Go

You: "What's a server?"
Teacher: "Imagine your website is a restaurant. The server is the 
         kitchen — it's where all the food (your web pages) gets 
         prepared before being served to customers (your visitors). 
         Does that make sense?"

What Makes This Different

Technical terms are explained, not hidden. When the agent says "server", it immediately adds "(the computer that runs your website)". You learn the words naturally without being overwhelmed.

Agents suggest, not ask. Non-devs don't know what to ask for. Instead of "Want me to check the settings?", the agent says "I'm going to check the settings next — because after this kind of change, they sometimes need updating."

Every action has a meaning. Non-devs care about results, not processes:

| What the agent does | What it means for you | |---|---| | "Committing and pushing to git" | "Saving your changes and sharing them with the team" | | "Creating a pull request" | "Asking the team to review and approve your changes" | | "Running tests" | "Checking that everything still works correctly" | | "Building the project" | "Preparing your app so it's ready to go live" | | "Deploying to production" | "Putting your app online so users can see it" | | "Rolling back" | "Going back to the previous version because the new one had issues" | | "Creating an issue" | "Writing down a problem or idea so the team can track it" |

Commands

npx xplain-like-im-5 install <pack>    # Install a persona
npx xplain-like-im-5 install --all     # Install all 5
npx xplain-like-im-5 list              # See what's installed
npx xplain-like-im-5 remove <pack>     # Remove one
npx xplain-like-im-5 reset             # Remove all

How It Works

Each persona is a .md file installed to ~/.claude/agents/. When you start Claude Code with --agent xli5-<name>, Claude reads the persona instructions and adapts its communication style. No code changes, no config, no dependencies.

License

MIT