Agent Threat Rules (ATR)
An open MIT detection-rule standard for AI-agent and MCP attacks (prompt injection, tool poisoning, context exfiltration), like Sigma or YARA for the agent layer, with OWASP LLM/Agentic and MITRE ATLAS mappings on each rule.
- from
- Detection Engineering, Prompt Injection
- added
- 2026-10-10
- likes
- 0
Detection Engineering › Detection Content & Signatures: “An open MIT detection-rule standard for AI-agent and MCP attacks (prompt injection, tool poisoning, context exfiltration), like Sigma or YARA for the agent layer, with OWASP LLM/Agentic and MITRE ATLAS mappings on each rule.”
Prompt Injection › Tools: “Open detection standard for AI agent threats (prompt injection, tool poisoning, MCP attacks, skill compromise), using Sigma/YARA-style YAML rules. 330 rules across 9 attack categories with full mapping to OWASP Agentic Top 10 (10/10), MITRE ATLAS (100/113), NIST AI RMF (100%)…”