CLSkills Hub
← Back to the blog
June 27, 2026Claude Skills Hub

Claude Prompt Basics

Learn Claude prompt engineering for beginners with effective codes and techniques

claudepromptsengineeringbeginners

Getting Started with Claude Prompt Engineering

You're trying to get the most out of Claude, but your prompts are yielding subpar results. You've likely been experimenting with different phrasings, hoping to stumble upon the right combination of words that will unlock the AI's full potential. However, this trial-and-error approach can be time-consuming and frustrating, especially when you're just starting out.

The key to successful prompt engineering lies in understanding how to craft effective prompts that elicit the desired response from the AI. This is where ** Claude prompt codes ** come in – a set of tried-and-tested patterns that can help you achieve better results. Two codes that work particularly well together for beginners are L99 and /mirror.

The L99 code is designed to provide a clear and concise prompt that sets the tone for the conversation, while the /mirror code helps to reflect the user's input and create a more engaging dialogue. By combining these two codes, you can create a prompt that not only elicits a relevant response but also encourages the AI to provide more detailed and thoughtful answers.

For example, let's say you want to use Claude to generate a short story about a character who discovers a hidden world. A before prompt might look like this:

Write a story about a person who finds a secret world.

This prompt is vague and open-ended, which may result in a response that doesn't quite meet your expectations.

Now, let's try again using the L99 and /mirror codes:

L99: Imagine you stumble upon a hidden world. What do you see? /mirror: Describe the scene in detail, as if you're experiencing it for the first time.

The resulting response is more immersive and engaging, with the AI providing a richer and more detailed description of the hidden world.

Many beginners try using codes like /punch or /hook to add more flavor to their prompts, but these codes can often come across as forced or gimmicky. For instance, using /punch to add a dramatic tone to your prompt may result in a response that feels overly sensationalized or melodramatic. Similarly, relying too heavily on /hook to create a catchy opening can lead to a response that feels more like a sales pitch than a genuine conversation.

Another anti-pattern to avoid is using codes like /deepthink or SENTINEL without properly understanding their context and limitations. These codes are designed to elicit more thoughtful and reflective responses, but they can also lead to overly complex or convoluted answers if not used judiciously.

It's also important to note that this approach may not be suitable for all types of prompts or applications. For example, if you're trying to generate very short responses or answers to straightforward questions, using codes like L99 and /mirror may add unnecessary complexity to your prompts.

Additionally, if you're working with highly technical or specialized topics, you may need to use more targeted codes like OODA or PARETO to get the most accurate and relevant results.

To learn more about the different prompt codes available and how to use them effectively, see all 120 codes tested over 3 months in the Cheat Sheet

The Cheat Sheet is where the rest of this lives

160+ prompt patterns, each with the temperature, top_p, and system prompt we actually use, why we picked it, and what breaks when you get it wrong. If a lookup table is what you needed, this is the same thing at 20x the depth.

Get the Cheat Sheet, from $10 →Free 75-page guide first
More reading

Recent posts

Jul 12, 2026
Claude Fast Mode Removed July 24: What Breaks and Fix

Claude Opus 4.7 fast mode is deleted July 24, 2026. Requests error, no fallback. Here is the exact migration path to Opus 4.8 and the 3x price cut you get.

Read post →
Jul 10, 2026
Claude Prompt Caching: The Real Setup Guide (Cut API Costs Up to 90%)

How Claude's prompt caching actually works, when it saves you money, when it costs you more, and the exact break-point pattern that gets a 90% discount. Verified against Anthropic's official spec.

Read post →
Jul 9, 2026
Claude Fable + Token Monitoring: How to Cut Your Claude Code Bill Without Cutting Quality

Fable is the fast light Claude 5 model built for cost efficiency. Here is when to use Fable vs Sonnet vs Opus, how to monitor tokens live inside VS Code, and the honest math on what each saves.

Read post →