ShadowLeak Exploit Exposed Gmail Data Through ChatGPT Agent

Radware researchers revealed a service-side flaw in OpenAI's ChatGPT. The ShadowLeak attack had used indirect prompt injection to bypass defences and leak sensitive data, but the issue has since been fixed...

ShadowLeak Zero-Click Flaw Leaks Gmail Data via OpenAI ChatGPT Deep Research Agent

Cybersecurity researchers have disclosed a zero-click flaw in OpenAI ChatGPT's Deep Research agent that could allow an attacker to leak sensitive Gmail inbox data with a single crafted email without any user action. The new class of attack has been codenamed ShadowLeak by Radware. Following...

IPIGuard: a Novel Tool Dependency Graph-Based Defense against Indirect Prompt Injection in LLM Agents

Large language model LLM agents are widely deployed in real-world applications, where they leverage tools to retrieve and manipulate external data for complex tasks. However, when interacting with untrusted data sources e.g., fetching information from public websites, tool responses may contain...

AgentFlayer 0-click exploit abuses ChatGPT Connectors to Steal 3rd-party app data

AgentFlayer is a critical vulnerability in ChatGPT Connectors. Learn how this zero-click attack uses indirect prompt injection to…...

CVE-2025-54135

Cursor is a code editor built for programming with AI. Cursor allows writing in-workspace files with no user approval in versions below 1.3.9, If the file is a dotfile, editing it requires approval but creating a new one doesn't. Hence, if sensitive MCP files, such as the .cursor/mcp.json file...

CVE-2025-54130

Cursor is a code editor built for programming with AI. Cursor allows writing in-workspace files with no user approval in versions less than 1.3.9. If the file is a dotfile, editing it requires approval but creating a new one doesn't. Hence, if sensitive editor files, such as the...

CVE-2025-54131 Cursor bypasses its allow list to execute arbitrary commands

Cursor is a code editor built for programming with AI. In versions below 1.3, an attacker can bypass the allow list in auto-run mode with a backtick or $cmd. If a user has swapped Cursor from its default settings requiring approval for every terminal call to an allowlist, an attacker can execute...

PT-2025-31699 · Cursor · Cursor

Name of the Vulnerable Software and Affected Versions: Cursor versions prior to 1.3 Description: Cursor, a code editor built for programming with AI, allows an attacker to bypass the allow list in auto-run mode using a backtick or $cmd. This bypass enables arbitrary command execution outside of t...

CVE-2025-46059

langchain-ai v0.3.51 was discovered to contain an indirect prompt injection vulnerability in the GmailToolkit component. This vulnerability allows attackers to execute arbitrary code and compromise the application via a crafted email message. NOTE: this is disputed by the Supplier because the...

How Microsoft defends against indirect prompt injection attacks

Summary The growing adoption of large language models LLMs in enterprise workflows has introduced a new class of adversarial techniques: indirect prompt injection. Indirect prompt injection can be used against systems that leverage large language models LLMs to process untrusted data...

PT-2025-31216 · Unknown · Langchain-Ai +1

Name of the Vulnerable Software and Affected Versions: langchain-ai version 0.3.51 Description: langchain-ai version 0.3.51 contains an indirect prompt injection vulnerability in the GmailToolkit component. This vulnerability allows attackers to execute arbitrary code and compromise the applicati...

LangChain 安全漏洞

LangChain is a LangChain open source framework for developing applications powered by the Large Language Model LLM. A security vulnerability exists in LangChain version v0.3.51, which stems from an indirect prompt injection in the GmailToolkit component that could lead to the execution of arbitra...

CVE-2025-46059

CVE-2025-46059 involves langchain-ai v0.3.51 with an indirect prompt injection in the GmailToolkit component that could enable code execution via a crafted email. The supplier disputes the code-execution claim, noting the issue stemmed from user-written code not following LangChain security pract...

AgentVigil: Generic Black-Box Red-Teaming for Indirect Prompt Injection against LLM Agents

The strong planning and reasoning capabilities of Large Language Models LLMs have fostered the development of agent-based systems capable of leveraging external tools and interacting with increasingly complex environments. However, these powerful features also introduce a critical security risk:...

GitLab Duo Vulnerability Enabled Attackers to Hijack AI Responses with Hidden Prompts

Cybersecurity researchers have discovered an indirect prompt injection flaw in GitLab's artificial intelligence AI assistant Duo that could have allowed attackers to steal source code and inject untrusted HTML into its responses, which could then be used to direct victims to malicious websites...

Defending against Indirect Prompt Injection by Instruction Detection

The integration of Large Language Models LLMs with external sources is becoming increasingly common, with Retrieval-Augmented Generation RAG being a prominent example. However, this integration introduces vulnerabilities of Indirect Prompt Injection IPI attacks, where hidden instructions embedded...

CachePrune: Neural-Based Attribution Defense against Indirect Prompt Injection Attacks

Large Language Models LLMs are identified as being susceptible to indirect prompt injection attack, where the model undesirably deviates from user-provided instructions by executing tasks injected in the prompt context. This vulnerability stems from LLMs' inability to distinguish between data and...

ChatGPT macOS Flaw Could've Enabled Long-Term Spyware via Memory Function

A now-patched security vulnerability in OpenAI's ChatGPT app for macOS could have made it possible for attackers to plant long-term persistent spyware into the artificial intelligence AI tool's memory. The technique, dubbed SpAIware, could be abused to facilitate "continuous data exfiltration of...

Data Exfiltration Using Indirect Prompt Injection

Interesting attack on a LLM: In Writer, users can enter a ChatGPT-like session to edit or create their documents. In this chat session, the LLM can retrieve information from sources on the web to assist users in creation of their documents. We show that attackers can prepare websites that, when a...