greshake/llm-security
Security research demonstrating indirect prompt injection vulnerabilities and attack techniques targeting integrated language models.
Not currently ranked — collecting fresh signals.
star history
This repository presents proof-of-concept demonstrations of a new class of vulnerabilities affecting LLMs integrated into applications. It shows how prompt injections can enable remote control of LLMs, exfiltration of user data, persistent compromise across sessions, and attacks on code completion engines like Copilot. The work is based on an academic paper published on ArXiv and includes working demos for GPT-4, GPT-3, and LangChain-based applications.
Frequently asked
- What is greshake/llm-security?
- Security research demonstrating indirect prompt injection vulnerabilities and attack techniques targeting integrated language models.
- Is llm-security open source?
- Yes — greshake/llm-security is open source, released under the MIT license.
- What language is llm-security written in?
- greshake/llm-security is primarily written in Jupyter Notebook.
- How popular is llm-security?
- greshake/llm-security has 2.1k stars on GitHub.
- Where can I find llm-security?
- greshake/llm-security is on GitHub at https://github.com/greshake/llm-security.