AI chat app leak exposes 300 million messages tied to 25 million users

An independent security researcher uncovered a major data breach affecting Chat & Ask AI, one of the most popular AI chat apps on Google Play and Apple App Store, with more than 50 million users. The researcher claims to have accessed 300 million messages from over 25 million users due to an...

Google Gemini Enterprise 安全漏洞

Google Gemini Enterprise is a generative AI platform developed by Google, Inc. of the United States. There is a security vulnerability in Google Gemini Enterprise, which stems from the use of predictable Google Cloud Storage bucket names. This vulnerability may allow attackers to preemptively tak...

Whispers of Wealth: Red-Teaming Google's Agent Payments Protocol Via Prompt Injection

Large language model LLM based agents are increasingly used to automate financial transactions, yet their reliance on contextual reasoning exposes payment systems to prompt-driven manipulation. The Agent Payments Protocol AP2 aims to secure agent-led purchases through cryptographically verifiable...

CVE-2026-0755

gemini-mcp-tool execAsync Command Injection Remote Code Execution Vulnerability. This vulnerability allows remote attackers to execute arbitrary code on affected installations of gemini-mcp-tool. Authentication is not required to exploit this vulnerability. The specific flaw exists within the...

GHSA-28QQ-5F47-R5X2 Duplicate Advisory: gemini-mcp-tool vulnerable to OS command injection and @file exfiltration via prompt quoting (CVE-2026-0755)

Duplicate Advisory This advisory has been withdrawn because it is a duplicate of GHSA-4h5r-5jm8-jxjm. This link is maintained to preserve external references. Original Description gemini-mcp-tool execAsync Command Injection Remote Code Execution Vulnerability. This vulnerability allows remote...

CVE-2026-0755

gemini-mcp-tool execAsync Command Injection Remote Code Execution Vulnerability. This vulnerability allows remote attackers to execute arbitrary code on affected installations of gemini-mcp-tool. Authentication is not required to exploit this vulnerability. The specific flaw exists within the...

CVE-2026-0755

gemini-mcp-tool execAsync Command Injection Remote Code Execution Vulnerability. This vulnerability allows remote attackers to execute arbitrary code on affected installations of gemini-mcp-tool. Authentication is not required to exploit this vulnerability. The specific flaw exists within the...

CVE-2026-0755

CVE-2026-0755 affects the gemini-mcp-tool, specifically the execAsync method. The flaw is an OS command injection (CWE-78) caused by insufficient validation of user-supplied input before it is used in a system call, enabling unauthenticated remote code execution. Attackers could run arbitrary cod...

CVE-2026-0755 gemini-mcp-tool execAsync Command Injection Remote Code Execution Vulnerability

gemini-mcp-tool execAsync Command Injection Remote Code Execution Vulnerability. This vulnerability allows remote attackers to execute arbitrary code on affected installations of gemini-mcp-tool. Authentication is not required to exploit this vulnerability. The specific flaw exists within the...

CVE-2026-0755 gemini-mcp-tool execAsync Command Injection Remote Code Execution Vulnerability

gemini-mcp-tool execAsync Command Injection Remote Code Execution Vulnerability. This vulnerability allows remote attackers to execute arbitrary code on affected installations of gemini-mcp-tool. Authentication is not required to exploit this vulnerability. The specific flaw exists within the...

Google Gemini MCP Tool operating system command injection vulnerability

Google Gemini MCP Tool is a tool component developed by Google Inc., based on large model context protocols. Google Gemini MCP Tool has a vulnerability related to operating system command injection. This vulnerability stems from the execAsync method, which executes system calls without verifying...

context_compaction

Google ADK Context Compaction POC A proof of concept demonstr...

Google Gemini Prompt Injection Flaw Exposed Private Calendar Data via Malicious Invites

Cybersecurity researchers have disclosed details of a security flaw that leverages indirect prompt injection targeting Google Gemini as a way to bypass authorization guardrails and use Google Calendar as a data extraction mechanism. The vulnerability, Miggo Security's Head of Research, Liad...

Google Gemini AI Tricked Into Leaking Calendar Data via Meeting Invites

Cybersecurity researchers at Miggo Security found a flaw in Google Gemini that uses calendar invites to steal private data. Learn how this silent attack bypasses security...

BIT-KIBANA-2026-0532 External Control of File Name or Path and Server-Side Request Forgery (SSRF) in Kibana Google Gemini Connector

External Control of File Name or Path CWE-73 combined with Server-Side Request Forgery CWE-918 can allow an attacker to cause arbitrary file disclosure through a specially crafted credentials JSON payload in the Google Gemini connector configuration. This requires an attacker to have authenticate...

BIT-ELK-2026-0532 External Control of File Name or Path and Server-Side Request Forgery (SSRF) in Kibana Google Gemini Connector

External Control of File Name or Path CWE-73 combined with Server-Side Request Forgery CWE-918 can allow an attacker to cause arbitrary file disclosure through a specially crafted credentials JSON payload in the Google Gemini connector configuration. This requires an attacker to have authenticate...

CVE-2026-0532

A flaw was found in Kibana. This vulnerability allows an authenticated attacker, with privileges to create or modify connectors, to disclose arbitrary files. The attacker achieves this by submitting a specially crafted configuration for the Google Gemini connector, which the server processes...

Kibana 8.x < 8.19.10 / 9.1.x < 9.1.10 / 9.2.x < 9.2.4 (ESA_2026_05)

The version of Kibana installed on the remote host is prior to 8.19.10, 9.1.10, or 9.2.4. It is, therefore, affected by a vulnerability as referenced in the ESA202605 advisory. - An external control of file name or path combined with a server-side request forgery SSRF vulnerability exists in the...

CVE-2026-0532

External Control of File Name or Path CWE-73 combined with Server-Side Request Forgery CWE-918 can allow an attacker to cause arbitrary file disclosure through a specially crafted credentials JSON payload in the Google Gemini connector configuration. This requires an attacker to have authenticate...

External Control of File Name or Path

Overview kibana is an open source Apache Licensed, browser-based analytics and search dashboard for Elasticsearch. Affected versions of this package are vulnerable to External Control of File Name or Path via the processing of JSON credentials in the Google Gemini connector configuration. An...