ADR
ADR secures enterprise AI agents through observability, security benchmarking, and threat detection. Deployed at Uber.
What is ADR?
ADR secures enterprise AI agents through observability, security benchmarking, and threat detection. Deployed at Uber.
Overview of ADR
What it is
ADR (Agentic AI Detection and Response) is an enterprise security system for AI agents. It is deployed in production at Uber and was accepted to MLSys 2026. The open-source release includes ADR Discovery, ADR Sensor, ADR-Bench, and ADR Detector.
Key features
- ADR Discovery: inventories installed AI applications, CLI agents, IDE extensions, local model runtimes, and MCP servers on endpoints, flagging unknown surfaces for review.
- ADR Observability: captures agent intent, tool use, and execution traces across 7+ AI coding tools on macOS, Linux, and Windows, as well as internal automation and customer-facing support agents.
- ADR Benchmark: includes 300+ tasks, 134 MCP servers, and coverage of all 17 agent attack techniques for testing agent security under realistic enterprise conditions.
- ADR Detection: uses a two-tier architecture combining high-recall triage with deeper agentic reasoning for suspicious sessions.
Use cases
Securing employee-facing AI agents such as Cursor, Claude Code, Codex, GitHub Copilot CLI, and DeepSeek Harness. Also securing customer-facing agents such as AI support agents.
Getting started
git clone https://github.com/uber/ADR
cd ADR/Detection
uv sync
export ANTHROPIC_API_KEY="..." OPENAI_API_KEY="..."
For keyless smoke tests, use --detector llamafirewall. See Detection/README.md and docs/REPRODUCIBILITY.md for full evaluation workflow.
License and pricing
Licensed under Apache License 2.0. Detection/benchmark/agentdojo/ is vendored third-party code under its own MIT license. The project is open source.
Based on the project's public README and repository metadata.