💀 Faced with a pozzed key? Claude avoids ERP? Here is the SOLUTION.

XMLK's Explanations

🔞SillyTavern enthusiast, Claude enjoyer, GPT struggler

Send a cute m/f kobold, feedback and suggestions! Email: [email protected] Discord: victorianmaids I read every email and try to implement your suggestions and findings in my prompts.

â €
â €

Be careful while using custom proxy builds.

â €
â €

~If you don't see images - the catbox is down!~

â €


ReadMe

BEFORE WE DIVE IN ANY FURTHER PLEASE READ THIS DISCLAIMER:

• I do not encourage/propagate any illegal/questionable/harmful/unethical activities or hate towards any ethnic groups/nationalities/minorities etc. All prompts and "jailbreaks" are intended for entertaining purposes ONLY.
• If you want to get access to a certain LLM so you can chat with bots, get access legally. When violating the Terms of use keep in mind that access can be taken from you by the key provider [Anthropic/OpenAI].
• Roleplays and generated stories depict exclusively fictional characters and scenarios, do not represent/encourage/propagate real experiences and are not intended for publishing. Private use only.
• My prompts are a result of work by dozens of AI RP enthusiasts, not just me. Sometimes I come up with something myself or take inspiration from the work of others, sometimes I copy parts of prompts from other 'prompters'. You may modify and repost my prompts too, no mentions needed.
• Using these prompts you are confirming that you are 18+ and in a clear state of mind.
• While RP with AI is fun, be mindful about how you spend your time. Do not get too attached to your characters. Don't forget about your real friends and close ones, hobbies, shower, walks.
• Be sure to take rests between khm sessions and mind your health. Stop if you are uncomfortable or feel pain/discomfort. I am really-really serious. Dead serious.





OUTDATED V8+




Why we use XML?

Quote from anthropic docs: "We can also ask Claude to use XML tags in its response. Doing so can make it easy to extract key information in a setting where the output is automatically processed. Claude is naturally very chatty, so requesting these output XML tags helps separate the response itself from Claude's comments on the response."

Claude was specifically trained to understand XML tags. Breaking your prompt in <sections> and explaining/repeating makes it easier for AI to follow your prompt!

Read Anthropic's documentation

These images are just VISUALISATIONS AND EXPLANATIONS on how the prompt works! All you need to do is explained in Version history / Main page for each version. YOU ONLY NEED TO COPY-PASTE CODE BLOCKS from renty!!!

Prompt structure

What can be safely edited to align with your own RP?

Simplified prompt chain visualization

Explaining each section

-Yep, there are spelling mistakes in images.
-Will I fix them?
-No. No, I don't think I will.

Edit
Pub: 07 Aug 2023 17:18 UTC
Edit: 16 Sep 2023 19:27 UTC
Views: 1310