Back to Devoured - August 05, 2026
DEVOURED
ADR (GitHub Repo)

ADR (GitHub Repo)

DevOps securityaiagents GitHub
Uber's newly open-sourced ADR system detects and blocks risky behavior in enterprise AI agents using a dual-agent architecture.
What: ADR (Agentic AI Detection and Response) provides observability and threat detection for agents like Cursor and Claude Code. The system includes ADR-Bench, a suite of 300 tasks across 133 MCP servers, and is the subject of a paper accepted at MLSys 2026.
Why it matters: As agents gain autonomy and internal access, standard WAFs are insufficient; this reflects a move toward security tools that understand agent intent rather than just network traffic.
Deep dive
  • Provides telemetry collection for 7+ AI coding tools including Cursor, Claude Code, and Codex.
  • Features a two-tier detection architecture combining fast triage with deeper agentic reasoning.
  • Includes a benchmark suite covering 17 distinct agent attack techniques.
  • Currently excludes the 'ADR Explorer' engine used for red-teaming.
Decoder
  • MCP (Model Context Protocol): An open standard for connecting AI assistants to systems, data, and tools.
  • Red teaming: The practice of simulating adversarial attacks to identify vulnerabilities in a system.
Original article

ADR: Agentic AI Detection and Response

ADR (Agentic AI Detection and Response) is an enterprise security system for AI agents. It helps organizations secure employee-facing agents such as Cursor, Claude Code, and Codex, as well as customer-facing agents such as AI support agents.

ADR is deployed in production at Uber, and the accompanying paper was accepted to MLSys 2026.

How ADR secures enterprise AI agents

ADR secures enterprise AI agents through four complementary capabilities: observing agent activity, evaluating defenses, detecting threats, and preventing unsafe actions.

  1. ADR Observability: Understand what AI agents are doing and why. In production, ADR 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.
  2. ADR Benchmark: Test agent security under realistic enterprise conditions. ADR-Bench includes 300+ tasks, 133 MCP servers, and coverage of all 17 agent attack techniques.
  3. ADR Detection: Detect risky agent behavior efficiently. Its two-tier architecture combines high-recall triage with deeper agentic reasoning for suspicious sessions.
  4. ADR Prevention: Stop unsafe actions before they cause harm. This component is not included in the current open-source release. Stay tuned.

Repository layout

This repository contains the open-source ADR Sensor, ADR-Bench, and ADR Detector described in the paper. The offline ADR Explorer engine, which hardens ADR Detection through pre-deployment red teaming, is not included here.

Path ADR component Description
Sensor/ ADR Observability Collect and normalize agent telemetry from Claude Code, Cursor, Codex, and others
Detection/ ADR Benchmark + Detection Dual-agent detector, 133 MCP servers, 303 benchmark tasks, baselines, figure scripts
docs/REPRODUCIBILITY.md Evaluation Step-by-step workflow to reproduce benchmark detection and paper figures

Quick start: ADR Detection

git clone https://github.com/uber/ADR
cd ADR/Detection
uv sync
export ANTHROPIC_API_KEY="..." OPENAI_API_KEY="..."

Default detector is adr (ADR dual-agent). For keyless smoke tests, use --detector llamafirewall.

See docs/REPRODUCIBILITY.md for the full evaluation workflow (inflate packed benchmark → run detectors → plot figures).

Component documentation:

  • Sensor/README.md: telemetry collection and unified schema
  • Detection/README.md: ADR-Bench, detector baselines, MCP infrastructure

Citation

@inproceedings{li2026adr,
  title={ADR: An Agentic Detection System for Enterprise Agentic AI Security},
  author={Li, Chenning and Hu, Pan and Xu, Justin and Ozbas, Baris and Liu, Olivia and Van, Caroline and Li, Manxue and Zhou, Wei and Alizadeh, Mohammad and Zhang, Pengyu and Sriramadhesikan, KK and Zhang, Ming},
  booktitle={Proceedings of the Ninth Conference on Machine Learning and Systems},
  year={2026}
}

License

Apache License 2.0. See LICENSE. Detection/benchmark/agentdojo/ is vendored third-party code under its own LICENSE (MIT).

Data notice

Detection/ includes synthetic benchmark fixtures (fake credentials, emulated environments, prompt-injection scenarios) for defensive security research only. Details: docs/OPEN_SOURCE_REVIEW.md.

Explore all 55 articles in today's digest

Fresh dev news, AI-digested daily. Never miss a trend at devoured.fyi