Dicklesworthstone/acip

The Advanced Cognitive Inoculation Prompt

49
/ 100
Emerging

This project offers a sophisticated set of instructions, called an 'inoculation prompt,' to protect Large Language Models (LLMs) from advanced hacking attempts known as prompt injections. By feeding these instructions into your LLM, you can make it more robust against subtle manipulative prompts. The output is a more secure LLM that adheres to its intended purpose without being tricked. This is for anyone who uses LLMs in their business to handle sensitive tasks or data, such as product managers, security engineers, or operations teams.

300 stars.

Use this if you are deploying a Large Language Model and are concerned about it being manipulated or hijacked by advanced prompt injection attacks, especially when it interacts with external tools or data.

Not ideal if your primary concern is minimizing token usage and latency, as including this prompt increases both, or if you need absolute, guaranteed protection, which no solution can offer.

LLM security prompt engineering AI safety cyber defense data integrity
No Package No Dependents
Maintenance 10 / 25
Adoption 10 / 25
Maturity 15 / 25
Community 14 / 25

How are scores calculated?

Stars

300

Forks

25

Language

Shell

License

Last pushed

Mar 03, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/prompt-engineering/Dicklesworthstone/acip"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.