🧰 Grok

The items that name it 35

• Leak Exposes 41,000 Homeland Security Workers–Including Undercover ICE Agents
  articleas classification.technologies2026-10-28
• Stolen AI Reasoning Could Help China Build Cheaper Rivals
  articleas classification.technologies2026-10-26
• Mikko Laakkonen (@mikkolaakkonen) on X
  threadas classification.technologies2026-10-17
• Frontier-Level Red-Teaming From a Self-Hostable Open-Weight Model: Testing NVIDIA Nemotron 3.5 Super
  articleas classification.technologies2026-10-15
• Can agents whose sole duty is to keep humans alive be hijacked to kill? Delayed multi-turn derived-action trap attacks in simulated high-stakes environments
  paperas classification.technologies2026-10-10
• My personal AI agent posted my bank details on company Slack
  articleas classification.technologies2026-10-10
• [AINews] not much happened today
  articleas classification.technologies2026-10-08
• OpenAI Says People Linked to China's Moonshot Tried to Copy Its AI's Hidden Reasoning
  articleas classification.technologies2026-10-01
• Huntress Tragic Quadrant: Top Cyber Threats Wrecking Businesses
  articleas affected products2026-10-01

  Cited as an example of a trusted AI domain that could be abused for AI platform abuse

• Gemini 4 Is Here, and Google’s Flagship Tops All Other AI Models on Cybersecurity
  articleas classification.technologies2026-09-30
• A Privacy Analysis of Web and Mobile Conversational AI Agents [pdf]
  paperas affected products2026-09-29It is written here as Grok (web: grok.com, Android: ai.x.grok).

  Tranco rank 956, >100M Play Store installs

• LLM Agents Can Easily Tamper With Their Own Traces
  paperas classification.technologies2026-09-24
• nealbridges/VulnHunter
  repoas classification.technologies2026-09-22
• Grok 4.7 for Offensive Security: Orchestration Matters | XBOW
  articleas classification.technologies2026-09-21
• ToolLeak: Tool-Invocation Attacks on Coding Agents
  paperas classification.technologies2026-09-21
• All-AI-Jailbreaks
  repoas affected products2026-09-17

  Model family targeted by collected jailbreak/prompt-injection experiments

• Top Agentic AI security resources — September 2026
  articleas affected products2026-09-04

  Target of zero-click 'Cryptographic Context Injection' data theft attack leaking chat histories, name, location, and subscription tier

• The AI Attack Surface: How Threat Actors Abuse Trusted AI Platforms
  writeupas classification.technologies2026-08-27
• Deep Eye
  repoas tools2026-08-22

  AI provider used for intelligent payload generation

• We burned 11.7bn tokens to find the best cyber AI model
  articleas classification.technologies2026-08-21
• Grok chat history leak: Cryptographic Context Injection
  writeupas classification.technologies2026-08-20
• XBOW: Grok 4.5 Is Powerful. The System Around It Makes It Safe.
  articleas classification.technologies2026-08-08
• System Prompts Leaks
  repoas affected products2026-08-03

  system prompts leaked, including older versions

• asgeirtj/system_prompts_leaks
  repoas affected products2026-08-03

  system prompts leaked, including older versions

• Kritt-ai/open-kritt
  repoas classification.technologies2026-08-02
• Benchmarking 13 AI models on rediscovering known CVEs
  articleas classification.technologies2026-07-16
• XBOW: Grok 4.5 Poised To Take Over the Middle of the AI Security Market
  articleas classification.technologies2026-07-16
• Destructive Command Guard (dcg)
  repoas affected products2026-07-15It is written here as Grok (xAI).

  Native ~/.grok/hooks/ support plus Claude compatibility layer; configurable via dcg install --grok

• HOL Guard
  repoas affected products2026-07-15
• Social Engineering Leveled Up. Has Your Security Program?
  articleas classification.key entities2026-05-01
• Attackers Didn’t Wait for AI. They Built Workflows Around It.
  articleas classification.technologies2026-04-22
• AI-Poisoning & AMOS Stealer: The Biggest Mac Threat | Huntress
  writeupas affected products2025-12-09It is written here as Grok (grok.com).

  Legitimate platform abused via SEO-poisoned shared conversation hosting malicious Terminal instructions, with a slightly different base64-encoded payload URL

• dfbench — depthfirst
  articleas classification.technologies
• SpecterOps/Blacklight — local AI agent attack surface mapping toolkit
  repoas affected products

  Supported by Scout for metadata-only reconnaissance; not yet supported by Session Analysis in this release

• AI Recommendation Poisoning: How
  articleas affected products

  supports deep-linked queries via URL parameters (grok.com/?q=...) exploited by the attack

In every workspace you can read 65