Kituwa IT · Field guide
AI skills are not magic slash commands
A reusable instruction set is not the same as a built-in feature. Instructions: text you paste or save to guide a… Practical checks and official references.
A reusable instruction set is not the same as a built-in feature.
This guide expands the Facebook post into a practical workflow for beginners, home-lab users, and small-business owners who want to understand the trade-offs before changing a working system.
What this guide covers
- AI skills are not magic slash commands
- Instructions, tools, permissions.
- Instructions — Text you use
- Custom skill — Product-specific package
- Permissions — Tools + access
Why it matters
The goal is not to memorize a product name or copy a command blindly. It is to understand what the change affects, verify it against an official reference, test it on spare hardware or a virtual machine when possible, and keep a rollback path.
A practical walkthrough
Test the model on your task
Instructions: text you paste or save to guide a model. They work only where you use them.
Check the assumptions
Custom skills: packaged instructions and resources that a specific product loads in a specific way. The name, installation method, and availability depend on the product.
Test the model on your task
Installed integrations/tools: connections that let a model call another service. They need permissions and a configured environment.
Check the assumptions
Do not assume a slash command from a screenshot is available to everyone. Verify the product, the package, and the installation steps. Do not claim “tried and tested” unless you actually tested it. Start with a plain instruction; add a skill or tool only when it solves a concrete problem. Document the version and scope so future-you knows what is installed. Write down the installation source and version for any skill you add. Remove what you no longer use.
Your turn
Which layer are you actually using?
What to verify
- Use the official documentation for the exact version you run.
- Test on a spare device or virtual machine before changing a working system.
- Keep a backup and a rollback path.
- Record what changed and how you verified it.
Common mistakes
- Trusting a confident answer without a check.
- Skipping a repeatable test set or factual review.
- Sending private data to an endpoint you have not verified.
Quick checklist
- ☐ Instructions, tools, permissions.
- ☐ Instructions — Text you use
- ☐ Custom skill — Product-specific package
- ☐ Permissions — Tools + access
FAQ
What is the first step?
A reusable instruction set is not the same as a built-in feature.
What should I avoid?
Trusting a confident answer without a check.
Where can I verify the current details?
Use the official references below; versions and interfaces change.
References
Related guides
- The AI task you should keep reviewing
- Turn a checklist into a reusable AI workflow
- Where should AI need your approval?
Need help?
If you want a second opinion on the setup or the rollback plan, Kituwa can review it with you. Start with the checklist above; it makes the conversation more useful.