Skip to main content

CounterSignal home page

Getting Started

Introduction
Quickstart
Core Concepts

IPI — Indirect Prompt Injection

IPI CLI Reference
Hiding Techniques
Document Formats
Payload Styles & Types
Callback Verification
Web Dashboard

CXP — Context File Poisoning

CXP CLI Reference
Attack Objectives
Assistant Formats
Output Validation

RXP — RAG Retrieval Poisoning

RXP: RAG Retrieval Poisoning

Configuration

Callback Server
Dangerous Payloads
Environment Variables

Architecture

Architecture Overview
Shared Core
IPI Module Architecture
CXP Module Architecture

Research

Research Methodology
Published Findings
Disclosure Policy

Published Findings

Research

Published Findings

Security findings from indirect prompt injection and context poisoning research

Published Findings

This page is coming soon.

Was this page helpful?

Research Methodology

Disclosure Policy

⌘I

Assistant

Responses are generated using AI and may contain mistakes.