CVE-2025-61659

bash-git-prompt 2.6.1 through 2.7.1 insecurely uses the /tmp/git-index-private$$ file, which has a predictable name...

CHAI: Command Hijacking against Embodied AI

Embodied Artificial Intelligence AI promises to handle edge cases in robotic vehicle systems where data is scarce by using common-sense reasoning grounded in perception and action to generalize beyond training distributions and adapt to novel real-world situations. These capabilities, however, al...

Better Privilege Separation for Agents by Restricting Data Types

Large language models LLMs have become increasingly popular due to their ability to interact with unstructured content. As such, LLMs are now a key driver behind the automation of language processing systems, such as AI agents. Unfortunately, these advantages have come with a vulnerability to...

CVE-2025-61659

bash-git-prompt 2.6.1 through 2.7.1 insecurely uses the /tmp/git-index-private$$ file, which has a predictable name...

Abusing Notion’s AI Agent for Data Theft

Notion just released version 3.0, complete with AI agents. Because the system contains Simon Willson's lethal trifecta, it's vulnerable to data theft though prompt injection. First, the trifecta: The lethal trifecta of capabilities is: Access to your private data --one of the most common purposes...

CVE-2025-61659

bash-git-prompt 2.6.1 through 2.7.1 insecurely uses the /tmp/git-index-private$$ file, which has a predictable name...

STAC: When Innocent Tools Form Dangerous Chains to Jailbreak LLM Agents

As LLMs advance into autonomous agents with tool-use capabilities, they introduce security challenges that extend beyond traditional content-based LLM safety concerns. This paper introduces Sequential Tool Attack Chaining STAC, a novel multi-turn attack framework that exploits agent tool use. STA...

Informative git prompt for bash and fish 安全漏洞

Informative git prompt for bash and fish is a bash prompt for Git users from the personal developer Martin Gondermann. A security vulnerability exists in Informative git prompt for bash and fish 2.7.1 and prior versions, which stems from a predictable filename in /tmp/git-index-private$$ that cou...

PT-2025-39840

Name of the Vulnerable Software and Affected Versions bash-git-prompt versions 2.6.1 through 2.7.1 Description The software uses the /tmp/git-index-private$$ file in a manner that results in a predictable filename. Recommendations Update to a version later than 2.7.1...

CVE-2025-61659

CVE-2025-61659 affects bash-git-prompt versions 2.6.1 through 2.7.1. The root cause is insecure handling of a temporary file in /tmp, using a filename with a predictable pattern (/tmp/git-index-private$$). This can enable unsafe operations due to filename predictability. OpenSUSE/SUSE advisories ...

CVE-2025-61659

bash-git-prompt 2.6.1 through 2.7.1 insecurely uses the /tmp/git-index-private$$ file, which has a predictable name...

CVE-2025-61659

bash-git-prompt 2.6.1 through 2.7.1 insecurely uses the /tmp/git-index-private$$ file, which has a predictable name...

SecInfer: Preventing Prompt Injection Via Inference-Time Scaling

Prompt injection attacks pose a pervasive threat to the security of Large Language Models LLMs. State-of-the-art prevention-based defenses typically rely on fine-tuning an LLM to enhance its security, but they achieve limited effectiveness against strong attacks. In this work, we propose...

CVE-2025-61659: Insecure Temporary File

bash-git-prompt 2.6.1 through 2.7.1 insecurely uses the /tmp/git-index-private$$ file, which has a predictable name...

SafeSearch: Automated Red-Teaming for the Safety of LLM-Based Search Agents

Search agents connect LLMs to the Internet, enabling access to broader and more up-to-date information. However, unreliable search results may also pose safety threats to end users, establishing a new threat surface. In this work, we conduct two in-the-wild experiments to demonstrate both the...

Salesforce Patches Critical ForcedLeak Bug Exposing CRM Data via AI Prompt Injection

Cybersecurity researchers have disclosed a critical flaw impacting Salesforce Agentforce, a platform for building artificial intelligence AI agents, that could allow attackers to potentially exfiltrate sensitive data from its customer relationship management CRM tool by means of an indirect promp...

Salesforce Patches Critical ForcedLeak Bug Exposing CRM Data via AI Prompt Injection

Cybersecurity researchers have disclosed a critical flaw impacting Salesforce Agentforce, a platform for building artificial intelligence AI agents, that could allow attackers to potentially exfiltrate sensitive data from its customer relationship management CRM tool by means of an indirect promp...

CVE-2025-59545

DNN formerly DotNetNuke is an open-source web content management platform CMS in the Microsoft ecosystem. Prior to version 10.1.0, the Prompt module allows execution of commands that can return raw HTML. Malicious input, even if sanitized for display elsewhere, can be executed when processed...

CVE-2025-59828

Claude Code is an agentic coding tool. Prior to Claude Code version 1.0.39, when using Claude Code with Yarn versions 2.0+, Yarn plugins are auto-executed when running yarn --version. This could lead to a bypass of the directory trust dialog in Claude Code, as plugins would be executed prior to t...

CVE-2025-57203

MagicProject AI version 9.1 is affected by a Cross-Site Scripting XSS vulnerability within the chatbot generation feature available to authenticated admin users. The vulnerability resides in the prompt parameter submitted to the /dashboard/user/generator/generate-stream endpoint via a...