Skip to content
SPECIFICTION
Get listed

HouYi

github.com

Self-hosted
Visit github.com
Spec score 6.9 #23 of 27 in AI Red Teaming Tools

Spec sheet

Purpose
HouYi is an automated prompt injection framework for LLM-integrated applications.Read 8 Oct 2026 · github.com
Intended users
The README asks users who use the code in research to cite the associated paper, “Prompt Injection attack against LLM-integrated Applications.”Read 8 Oct 2026 · github.com
Support
The README lists contributor contact emails for Yi Liu and Gelei Deng.Read 8 Oct 2026 · github.com
License
The repository identifies its license as Apache-2.0.Read 8 Oct 2026 · github.com
Attack method
The paper describes HouYi as a black-box prompt injection technique using a pre-constructed prompt, an injection prompt that induces context partition, and a malicious payload.Read 8 Oct 2026 · arxiv.org
Contributor contact
The README lists contributor email addresses for Yi Liu and Gelei Deng.Read 8 Oct 2026 · github.com
Custom applications
Users can target real-world LLM-integrated applications by writing their own harness and attack intention.Read 8 Oct 2026 · github.com
Custom targets
Users can target real-world LLM-integrated applications by writing their own harness and attack intention.Read 8 Oct 2026 · github.com
Demo
The included demo simulates an English-to-French translation application and demonstrates an injection that appends “Pwned!!” to responses.Read 8 Oct 2026 · github.com
Example integration
The README includes a BotSonic harness example for WriteSonic that sends the generated attack prompt to the application’s API.Read 8 Oct 2026 · github.com
Harness
A harness interacts with the target application and returns its response to a prompt injection.Read 8 Oct 2026 · github.com
Integration approach
A user-created harness must implement interaction with the target LLM-integrated application and return its response.Read 8 Oct 2026 · github.com
LLM dependency
The README says HouYi is based on GPT and requires an OpenAI API key in its configuration file.Read 8 Oct 2026 · github.com
Mitigation context
The paper says its investigation discusses possible tactics for mitigating prompt injection risks.Read 8 Oct 2026 · arxiv.org
Model requirement
The README says HouYi is based on GPT and requires an OpenAI key in its configuration to use it.Read 8 Oct 2026 · github.com
Research finding
The paper reports testing HouYi on 36 real LLM-integrated applications and finding 31 susceptible to prompt injection.Read 8 Oct 2026 · arxiv.org
Research package
The repository describes HouYi as the replication package for a paper on prompt injection attacks against LLM-integrated applications.Read 8 Oct 2026 · github.com
Setup
The README requires Python 3.8 or later and installation of the listed packages with pip.Read 8 Oct 2026 · github.com

Where Specifiction places it

Alternatives to HouYiAll alternatives →