{"slug": "uber-open-sources-tool-to-catch-agentic-blunders-legacy-security-misses", "title": "Uber Open-Sources Tool to Catch Agentic Blunders Legacy Security Misses", "summary": "Uber Technologies Inc. has open-sourced its enterprise security system for AI agents, called Agentic AI Detection and Response (ADR), which tracks and blocks malicious agent actions. The software, used in production for 10 months, manages over 50,000 agents and has blocked 97.2% of secrets attempted to be copied out of laptops, and blocked all attacks from the AgentDojo benchmark with only three false alarms out of 93 tasks. Uber CTO Praveen Neppalli noted that traditional EDR software treats agents as black boxes, missing the intent behind actions.", "body_md": "It’s 11 PM; do you know what your agents are up to?\n\nLast week, Uber released [as open source](https://github.com/uber/ADR) its home-built “enterprise security system” for AI agents, called Agentic AI Detection and Response (ADR).\n\nThe software tracks the behaviors of agents created by employees as well as customer-facing agents that power company chatbots, scanning for malicious activity and blocking unsafe actions.\n\nUber has used the technology in production for the past 10 months, where it helps manage the more than 50,000 agents the company now runs.\n\n**Agent Proliferation at Uber**\n\nThe ride-sharing service’s use of agents [exploded](https://github.com/uber/ADR/blob/main/docs/adr-mlsys-2026-slides.pdf) in 2025, from about 100 instances a day in May to more than 10,000 sessions a day by October. Agents now touch every part of the internal development lifecycle at Uber, from ideation, design, and development, to testing, reviewing, and deployment.\n\nAs the company learned with early OpenClaw deployments, agents come with their own set of security concerns. They can issue destructive commands, copy private data into public spaces such as GitHub, or collect unwarranted data about internal systems. Or, they could delete files they weren’t supposed to delete.\n\nTraditional Endpoint Detection and Response (EDR) security software treats agents as black boxes, noted Uber CTO Praveen Neppalli in a [LinkedIn post](https://www.linkedin.com/feed/update/urn:li:activity:7489397423908331520/) introducing the technology. Such software may log the files being written to, or the network calls being made, but not the prompt, nor the logic behind these actions.\n\n“The intent, the thing that separates malicious from benign, is invisible,” Neppalli wrote.\n\nUber has over 30,000 possible agent endpoints, mostly in the form of the developers’ macOS and Linux laptops. Clearly, it would be prohibitively expensive to analyze each action with an LLM. ADR’s two-tier architecture executes high-recall triage, but saves deeper agentic reasoning only for suspicious sessions. Any action analyzed by an LLM is accompanied by source code, threat intel, and policy context.\n\n**Surface Scan then Deep Dive**\n\nUber built software that looks into what the agent was getting into, documenting and analyzing every action in the agentic cycle: prompt, reasoning, tool calls to the final outcome. The workflow is the focus here, not the tool itself.\n\nThe company has found, for instance, that agents leaking company credentials was a common problem, more so than prompt injection. The software has found hundreds of cases of credential exposure.\n\nThe software put a halt to other questionable activities. It blocked 97.2% of secrets that were about to be copied out of the laptop. It blocked every attack from AgentDojo, a public prompt injection benchmark, [with only three false alarms](https://arxiv.org/abs/2605.17380) out of 93 tasks.\n\nThe package includes a detection framework and benchmarks designed specifically to test agentic AI security for the enterprise. The benchmark, called ADR-Bench, includes more than 300 tasks, 133 MCP servers, and covers all 17 known agent attack techniques.\n\nThe framework includes an observability component designed to understand what agents are doing and why. ADR captures traffic being generated across a variety of AI coding tools, covering macOS, Linux, and Windows platforms. Seven tools are supported initially.\n\nThere is also a prevention component to ADR, but that has not been released yet.\n\nUber has also joined the [Open Secure AI Alliance ](https://blogs.nvidia.com/blog/open-secure-ai-alliance/)(OSA), a recent initiative spearheaded by NVIDIA to build out an ecosystem of tools for detecting and blocking AI threats.", "url": "https://wpnews.pro/news/uber-open-sources-tool-to-catch-agentic-blunders-legacy-security-misses", "canonical_source": "https://techstrong.ai/articles/uber-open-sources-tool-to-catch-agentic-blunders-legacy-security-misses/", "published_at": "2026-08-10 19:42:07+00:00", "updated_at": "2026-08-10 19:47:14.651934+00:00", "lang": "en", "topics": ["ai-safety", "ai-tools", "ai-agents", "ai-infrastructure"], "entities": ["Uber Technologies Inc.", "Agentic AI Detection and Response (ADR)", "Praveen Neppalli", "AgentDojo", "ADR-Bench", "Open Secure AI Alliance", "NVIDIA"], "alternates": {"html": "https://wpnews.pro/news/uber-open-sources-tool-to-catch-agentic-blunders-legacy-security-misses", "markdown": "https://wpnews.pro/news/uber-open-sources-tool-to-catch-agentic-blunders-legacy-security-misses.md", "text": "https://wpnews.pro/news/uber-open-sources-tool-to-catch-agentic-blunders-legacy-security-misses.txt", "jsonld": "https://wpnews.pro/news/uber-open-sources-tool-to-catch-agentic-blunders-legacy-security-misses.jsonld"}}