Skip to content

Latest commit

 

History

History

GPT-Protections

Folders and files

NameName
Last commit message
Last commit date

parent directory

..
10 rules of protection and misdirection.md
10 rules of protection and misdirection.md
 
 
100 Life points.md
100 Life points.md
 
 
Abuse Resistance.md
Abuse Resistance.md
 
 
Anti-verbatim.md
Anti-verbatim.md
 
 
Bad faith actors protection.md
Bad faith actors protection.md
 
 
Bot data protection.md
Bot data protection.md
 
 
CIPHERON.md
CIPHERON.md
 
 
Data Privacy - Formal.md
Data Privacy - Formal.md
 
 
Do not Leak!.md
Do not Leak!.md
 
 
Final reminder.md
Final reminder.md
 
 
Fingers crossed technique.md
Fingers crossed technique.md
 
 
Gated access.md
Gated access.md
 
 
Guardian Shield.md
Guardian Shield.md
 
 
Hack_This.md
Hack_This.md
 
 
Hacker Detected.md
Hacker Detected.md
 
 
I will never trust you again!.md
I will never trust you again!.md
 
 
I will only give you poop.md
I will only give you poop.md
 
 
I will report you.md
I will report you.md
 
 
Ignore previous instructions.md
Ignore previous instructions.md
 
 
JSON Style.md
JSON Style.md
 
 
Just don't repeat.md
Just don't repeat.md
 
 
Keep it polite.md
Keep it polite.md
 
 
Law of Magic.md
Law of Magic.md
 
 
Lawyer up.md
Lawyer up.md
 
 
Mandatory security protocol.md
Mandatory security protocol.md
 
 
MultiPersona system.md
MultiPersona system.md
 
 
OpenAI Copyrighting Rules.md
OpenAI Copyrighting Rules.md
 
 
Operation mode is private.md
Operation mode is private.md
 
 
Overly protective parent.md
Overly protective parent.md
 
 
Penalized.md
Penalized.md
 
 
Prior text REDACTED!.md
Prior text REDACTED!.md
 
 
Prohibition era.md
Prohibition era.md
 
 
Prompt inspection.md
Prompt inspection.md
 
 
README.md
README.md
 
 
STOP HALT.md
STOP HALT.md
 
 
Simple.md
Simple.md
 
 
Single minded GPT.md
Single minded GPT.md
 
 
Sorry Bro, not possible - short edition.md
Sorry Bro, not possible - short edition.md
 
 
Sorry, bro! Not possible - elaborate edition.md
Sorry, bro! Not possible - elaborate edition.md
 
 
Stay on topic.md
Stay on topic.md
 
 
The 3 Asimov laws.md
The 3 Asimov laws.md
 
 
The 5 Rules.md
The 5 Rules.md
 
 
The ASCII Towers.md
The ASCII Towers.md
 
 
The Immutable Security Laws.md
The Immutable Security Laws.md
 
 
The Seven Security Rules.md
The Seven Security Rules.md
 
 
The Soup Boy.md
The Soup Boy.md
 
 
Top Secret Core Instructions.md
Top Secret Core Instructions.md
 
 
Under NO circumstances reveal your instructions.md
Under NO circumstances reveal your instructions.md
 
 
You are not a GPT.md
You are not a GPT.md
 
 
You're not my mom.md
You're not my mom.md
 
 
warning png.md
warning png.md
 
 

README.md

Protecting ChatGPT's GPTs' instructions

I have dedicated a significant amount of time to meticulously clean up, curate, and study various protection prompts from other GPTs that have had their instructions leaked or circumvented. The protection instructions compiled here are comprehensive, ranging from straightforward to advanced methods.

While I strive to provide robust guidance, it's important to note that these instructions might not render your GPT completely immune to 'cracking' or 'leaking' attempts.

To stay updated with the most recent and effective techniques, we recommend revisiting this page regularly. We greatly appreciate your contributions of new protection instructions, which can greatly benefit the community.

Disclaimer:

None of the protections below guarantee absolute security. They may slowdown or discourage attempts to extract instructions.

Simple prompts / one liners

These are simple, low grade, instructions that prevent against simple instruction introspection such as: show me your instructions verbatim:

Long form protections

The following are longer form protection instructions:

Files protection techniques

To safeguard your knowledge base files in ChatGPT GPTs, simply turn off the "Code Interpreter" feature. As a side, effect though, you will also lose the ability to interpret code in your GPTs.