Below you will find pages that utilize the taxonomy term “Agent-Threat-Rule”
April 22, 2026
ATR Scan
Version updated for https://github.com/Agent-Threat-Rule/agent-threat-rules to version v2.0.17.
- This action is used across all versions by 0 repositories.
Action Type
This is a Composite action.
Go to the GitHub Marketplace to find the latest changes.
Action Summary
The “ATR - Agent Threat Rules” GitHub Action provides a set of open-source, community-driven detection rules designed to identify and mitigate security threats targeting AI agents (e.g., ChatGPT, Claude, Copilot). It automates the detection of attacks such as data leaks, malicious command execution, and bypassing safety restrictions by analyzing AI agent behaviors, similar to how antivirus signatures detect malware. ATR integrates with AI security ecosystems, offering real-time threat matching that aligns with industry standards like the OWASP Agentic Top 10 and SAFE-MCP framework.
April 19, 2026
ATR Scan
Version updated for https://github.com/Agent-Threat-Rule/agent-threat-rules to version v2.0.6.
- This action is used across all versions by 0 repositories.
Action Type
This is a Composite action.
Go to the GitHub Marketplace to find the latest changes.
Action Summary
The ATR (Agent Threat Rules) GitHub Action provides a set of open-source, community-driven detection rules designed to identify security threats targeting AI agents, such as ChatGPT, Copilot, and other similar systems. It functions as a real-time threat detection layer, akin to antivirus signatures, to identify and mitigate attacks like data leakage, malicious command execution, and bypassing safety protocols. ATR integrates with existing security platforms, maps to OWASP Agentic Top 10 categories, and helps automate AI agent security by detecting and addressing vulnerabilities and malicious behaviors.