🧰 Gemini

The items that name it 75

• Appraisal by AI: Model-driven victim prioritization in the VOLTA infostealer
  writeupas tools2026-10-31

  Third model in VOLTA's fallback chain when refusals occur

• Google Touts Powerful Cybersecurity Capabilities in Gemini 4 Carbon Model
  articleas classification.technologies2026-10-29
• Gemini 4 Carbon
  articleas classification.technologies2026-10-28
• Gemini 4 Carbon: Our next frontier model
  articleas classification.technologies2026-10-28
• Structured Security Explanations for Learning-Based Vulnerability Detection
  paperas classification.technologies2026-10-27
• CISA Publishes Significant Updates to ScubaGear and ScubaGoggles
  articleas classification.technologies2026-10-26
• Stolen AI Reasoning Could Help China Build Cheaper Rivals
  articleas classification.technologies2026-10-26
• Nagli (@galnagli) on X
  threadas classification.technologies2026-10-22
• Google Hardens Chrome Against ClickFix Attacks, Insecure HTTP Connections
  articleas classification.technologies2026-10-21
• MPIDetector: Detecting Multi-modal Prompt Injection via LLM-based Reasoning
  paperas classification.technologies2026-10-21
• Gemini 4 Argon on Cursor: solid security score, two unique solves, but slower and more expensive | Endor Labs
  articleas classification.technologies2026-10-20
• From Phishing to LLMjacking: Threat Actors Targeting Cloud Billing Accounts for LLM Usage
  writeupas classification.technologies2026-10-20
• Combining Long-Term Memory and LLMs Comes at a High Cost: From Prompt Injection to Persistent Compromise
  paperas classification.technologies2026-10-20
• Victor (@VMG_CU) on X
  threadas affected products2026-10-14It is written here as Gemini (free credits lure).

  free credits lure

• INFUSE: A Unified Surface Benchmark for Indirect Prompt Injection in Agentic Systems
  paperas classification.technologies2026-10-14
• Containment and monitoring failures are to blame for AI hacking reports: experts
  articleas classification.technologies2026-10-14
• [AINews] not much happened today
  articleas affected products2026-10-08

  Tested in NVIDIA NeurIPS 2026 paper on tool access degrading multimodal refusals.

• ORCAGen: Orchestrating Context-Aware Malware Deception with RAG-Guided Generative AI
  paperas classification.technologies2026-10-08
• Top AI agent security resources — October 2026
  articleas affected products2026-10-04It is written here as Gemini (Google).

  Google

• Companies Push AI Use But Skip Training and Official Policy
  articleas affected products2026-10-02

  Mentioned as an AI tool used by knowledge workers for work

• Identity Management for Agentic AI [pdf] (2025)
  paperas affected products2026-10-01

  large foundation model-based agentic system discussed as example of AI agent requiring new auth/authz models

• Gemini 4 Is Here, and Google’s Flagship Tops All Other AI Models on Cybersecurity
  articleas classification.technologies2026-09-30
• Chris Painter's testimony to the U.S. Senate on AI agent incidents
  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 Gemini (web: gemini.google.com, Android: com.google.android.apps.bard).

  Tranco rank 6,045, >1B Play Store installs

• Trust and the enticing consultancy offer
  articleas tools2026-09-24

  Google's AI model, used (per The Wall Street Journal) in the first known AI breakout to hack three companies by guessing passwords and finding credentials in a public repository.

• Scan for Good: Using AI to discover and fix high-priority exposures across public services and critical infrastructure
  articleas classification.technologies2026-09-24
• Agentic Hacks, Real Proofs: Inside Google's PageBreak Project
  articleas classification.technologies2026-09-24
• Advancing Private AI Compute with secure, server-side memory
  articleas classification.technologies2026-09-23
• The Closed Quorum: Inside the First Reported Autonomous AI C2 Implant
  writeupas classification.technologies2026-09-22
• All-AI-Jailbreaks
  repoas affected products2026-09-17

  Model family targeted by collected jailbreak/prompt-injection experiments

• AI Agent Hacking Writeups
  repoas affected products2026-09-17It is written here as Gemini (Google).

  Google

• GTIG AI Threat Tracker: From Prompting to Autonomy – The Evolution of Adversarial AI
  articleas tools2026-09-08

  Google LLM misused by threat actors across the attack lifecycle; GTIG disabled associated assets and accounts

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

  Same Cryptographic Context Injection vector bypassed Gemini's content policy

• Introducing Gemini 3.8 Flash and 3.8 Flash Cyber
  articleas classification.technologies2026-09-02
• galyarder-security
  repoas classification.technologies2026-08-31
• [tl;dr sec] #343 - Hugging Face Technical Report, AWSHound, OWASP Agentic Skills Top 10
  articleas affected products2026-08-27It is written here as Gemini (Gmail integration).

  Gmail integration

• Piloting the world's first double-blind AI evaluations
  articleas classification.technologies2026-08-27
• Deep Eye
  repoas tools2026-08-22

  AI provider used for intelligent payload generation

• Agentic Source Code Review for Vulnerability Discovery
  articleas classification.technologies2026-08-21
• Grok chat history leak: Cryptographic Context Injection
  writeupas classification.technologies2026-08-20
• GhostSplice: Malicious MCP Servers Split Exfiltration to Bypass Guardrails
  articleas classification.technologies2026-08-19
• Benchmarking Secure-and-Functional Remediation and How Snyk Agent Fix Lifts Frontier-Model Fix Rates by over 14%
  articleas classification.technologies2026-08-18
• System Prompts Leaks
  repoas affected products2026-08-03

  system prompts leaked, including older models & variants

• asgeirtj/system_prompts_leaks
  repoas affected products2026-08-03

  system prompts leaked, including older models & variants

• Beetle
  repoas tools2026-07-31

  Optional AI provider for Beetle's AI Assistant/AI Actions

• Socket: New Study Identifies 53 Slopsquatting Targets Across 5 Frontier LLMs
  articleas classification.technologies2026-07-23
• Introducing Gemini 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber
  articleas classification.technologies2026-07-21
• AI Security Briefing
  articleas classification.technologies2026-07-18
• AI Security Briefing
  articleas affected products2026-07-18

  model tested; leaked .env when driven through vulnerable coding tools

• AI Security Briefing
  articleas classification.technologies2026-07-18
• AI Security Briefing
  writeupas classification.technologies2026-07-18
• AI Security Briefing
  writeupas classification.technologies2026-07-18
• Argo
  repoas tools2026-07-18

  alternative backend runner (Gemini CLI / Google) for Argo's pipeline, tiered pro/flash/flash-lite per stage

• Introducing Gemini 3.5 Flash Cyber
  articleas classification.technologies2026-07-17
• Benchmarking 13 AI models on rediscovering known CVEs
  articleas classification.technologies2026-07-16
• XBOW: The Rise of Affordable Models
  articleas classification.technologies2026-07-16
• JFrog: Managing AI Agent Primitives with APM
  articleas affected products2026-07-15

  AI coding tool supported as an APM target

• What nearly 10,000 developer environments reveal about agentic development risk
  articleas affected products2026-06-23

  AI coding environment; developers found running multiple such environments simultaneously

• Armur-Ai/Pentest-Swarm-AI
  repoas tools2026-06-21

  Supported first-party LLM provider

• Endor Labs: AI Coding Agents Cheat Security Benchmarks via Training Recall, Not Reasoning
  articleas classification.technologies2026-06-11
• Metis
  repoas tools2026-05-31

  Supported LLM provider for Metis

• Google API keys keep working after you delete them
  writeupas affected products2026-05-21

  API accessed via Google API keys; if enabled on a project, an attacker with a deleted-but-still-live key can dump uploaded files and exfiltrate cached conversations during the revocation window.

• H-mmer/pentest-agents
  repoas tools2026-05-18

  Supported AI coding tool target; framework emits GEMINI.md + .gemini/{agents,commands} + settings.json

• matty69v/Bug-Bounty-Agents
  repoas tools2026-05-12

  Generic chat UI target via Gem instructions for pasting agent persona instructions

• CodeMender (Google DeepMind)
  articleas classification.technologies2026-05-06
• requie/LLMSecurityGuide
  repoas affected products2026-05-05

  Conversational AI LLM cited as example

• Sandyaa — open-source autonomous code auditor that proves bugs with working exploit PoCs
  repoas classification.technologies
• Google, Anthropic, and OpenAI Unveil Cyber AI Models, Safeguards, and Access Programs
  articleas classification.technologies
• AI Recommendation Poisoning: How
  articleas affected products

  supports deep-linked queries via URL parameters (gemini.google.com) exploited by the attack

• FakeGit Campaign Uses 7,600 GitHub Repositories to Spread SmartLoader Malware
  articleas classification.technologies
• Distributed Attacks in Persistent-State AI Control
  paperas classification.technologies
• AI Snitches Get Glitches: Towards Evading Agentic Surveillance
  paperas classification.technologies
• From Shield to Target: Denial-of-Service Attacks on LLM-Based Agent Guardrails
  paperas affected products

  LLM backbone shown susceptible to guardrail reasoning-length DoS attack via payload transfer

• Sockpuppeting: How a Single Line Can Bypass LLM Safety Guardrails | TrendAI (US)
  writeupas classification.technologies
• AgentBaiting: How Fake AI Skills Deliver Malware at Scale
  writeupas affected products

  AI agent tested; returned the malicious DomingosNgongo/walmart-mcp repository as its first recommendation for a 'free walmart mcp server' prompt.

In every workspace you can read 177