Skip to content

[GitHub Trending] uber/ADR

7.9 relevance
Score Breakdown
technical depth
8
novelty
9
actionability
7
community
5
strategic
8
personal
10

Scored daily by a customisable AI persona to surface the most relevant engineering leadership news.

Enterprise AI agent security with observability and threat detection directly matches AI agent and security interests.

AI/ML github.com
ADR secures enterprise AI agents through observability, security benchmarking, and threat detection. Deployed at Uber. - uber/ADR
Summary

Uber open-sourced ADR (Agentic AI Detection and Response), an enterprise security system for AI agents deployed in production and accepted to MLSys 2026. It provides observability across 7+ AI coding tools on macOS, Linux, and Windows, a benchmark with 300+ tasks and 133 MCP servers covering 17 attack techniques, and a two-tier detection architecture combining high-recall triage with deeper agentic reasoning. The open-source release includes the Sensor, ADR-Bench, and Detector components, while the Prevention and Explorer modules remain proprietary.

Author

uber