ZeroHour
DashboardNewsCVEsExploitsIncidentsAIBriefAnalyticsEntitiesWatchlistIOCsAsk the feedSourcesFeeds & APIKiosk mode
Product

Agent Development Kit

0 mentions in 7 days · 2 in 30 days · 2 total · first seen · last

Search everything

Timeline

Google’s AI security agents found 100+ critical software vulnerabilities in just two days

Google Mandiant's AVDH, a chain of AI agents, found over 100 verified high-severity vulnerabilities and 12 assigned CVEs scanning code for ten months.

Google Mandiant disclosed AVDH (Agentic Vulnerability Discovery Harness), an internal pipeline of chained AI agents built on the Agent Development Kit that hunts vulnerabilities in source code. In a live investigation of stolen corporate repositories it verified more than 100 high-severity flaws in two days; over ten months it scanned tens of millions of lines of code and produced tens of thousands of findings, yielding 12 assigned CVEs including CVE-2026-13242 and CVE-2026-55803, with about a dozen more in active disclosure. Human consultants manually reproduce every confirmed finding before it counts.

Help Net Security · 27d agoToolsCVE-2026-13242CVE-2026-558031

Google’s $10,000 refund test shows why AI agents need zero trust

Google released an open-source zero-trust reference architecture for AI agents defending against prompt injection via signed database writes, gVisor sandboxing, and deterministic gating.

Google's demonstration, built on the Agent Development Kit (ADK) and Gemini, shows a customer support agent that could be manipulated into issuing a $10,000 refund on a $149 order and exposing environment variables via AI-generated Python. The architecture adds three security layers outside the model: cryptographic signatures on database writes verified via Cloud KMS backed by Cloud HSM, gVisor sandboxing of generated code with network egress disabled, and a Semantic Gateway applying deterministic checks to prompts and tool calls. It treats system prompts as insufficient boundaries because prompt injection, prompt tuning, or model updates can change agent behavior.

Help Net Security · 28d agoAI safety & security

Related CVEs

  • CVE-2026-132426.5<1%
    Improper Neutralization of Special Elements used in an SQL Command ('SQL Injection') vulnerability in Drupal Geolocation Field allows SQL Injection.
    Improper Neutralization of Special Elements used in an SQL Command ('SQL Injection') vulnerability in Drupal Geolocation Field allows SQL Injection. This issue affects Geolocation Field versions: from 0.0.0 to 3.15.0.
    · dopry geolocation field—
  • CVE-2026-558035.9<1%
    Improperly Controlled Modification of Dynamically-Determined Object Attributes vulnerability in Drupal Drupal core allows Object Injection.
    Improperly Controlled Modification of Dynamically-Determined Object Attributes vulnerability in Drupal Drupal core allows Object Injection. This issue affects Drupal core versions: from 0.0.0 to 10.5.12, from 10.6.0 to 10.6.11, from 11.2.0 to 11.2.14, from 11.3.0 to 11.3.12, from 0.0.0 to 11.0.*, from 0.0.0 to 11.1.*.
    · drupal drupal—

Appears with

organization Google 2vendor Google 2organization Mandiant 1ai model Gemini 1product Cloud HSM 1product Cloud KMS 1product gVisor 1product AVDH 1organization Google Threat Intelligence Group 1
Entities are extracted by the model from each article. Watching an entity keeps it in this browser only (no account); the watchlist page and dashboard alerts use it.