Common Vulnerabilities and Exposures ID
PulseAugur coverage of Common Vulnerabilities and Exposures ID — every cluster mentioning Common Vulnerabilities and Exposures ID across labs, papers, and developer communities, ranked by signal.
13 day(s) with sentiment data
-
AI assistants hallucinate package names, creating malware risk
An AI coding assistant recommended a non-existent package name to an engineer, a phenomenon known as "slopsquatting." The engineer narrowly avoided installing malware because they manually reviewed the package's registr…
-
New CTIFoundry corpus scaffold boosts LLM agent cyber threat intelligence
Researchers have introduced CTIFoundry, a new corpus scaffold designed to enhance the capabilities of LLM agents in cyber threat intelligence (CTI). This system structures CTI data by creating an ontology graph from aut…
-
New HARP Framework Prioritizes Vulnerabilities Using Adaptive LLM Fusion
Researchers have developed HARP, a novel framework for prioritizing Common Vulnerabilities and Exposures (CVEs) based on user preferences. Unlike existing systems that assume fixed criteria, HARP adapts to implicit oper…
-
Cybersecurity faces 'exposure gap' as AI accelerates vulnerability exploitation
The cybersecurity landscape is shifting, with attackers increasingly exploiting vulnerabilities before patches are even available, creating a significant "exposure gap." While visibility into exposures is improving, the…
-
New VICBench benchmark evaluates code vulnerability detection tools · 2 sources tracked
Researchers have introduced VICBench, a new benchmark designed to evaluate code vulnerability detection tools. The benchmark comprises 100 verified vulnerability-inducing commits (VICs) across 88 projects in Python, Jav…
-
AI-powered Eclipse plugin enhances Java code comprehension
Comprendia is a new Eclipse plugin designed to enhance Java code comprehension by integrating structural dependency visualization with LLM-powered explanations. The tool features a multi-edge-type dependency graph, LLM …
-
AI agent URL fetching vulnerability fixed with new tool
A developer has created a new tool called safe-fetch-mcp-server to address a critical security vulnerability in AI agents that fetch URLs. This vulnerability, known as Server-Side Request Forgery (SSRF), allows maliciou…
-
CyberForge framework generates synthetic data to train cybersecurity LLM agents
Researchers have developed CyberForge, a novel framework designed to generate synthetic security training data for large language model (LLM) agents. This framework injects verified vulnerabilities into real C/C++ softw…
-
Code models miss 80% of security fixes, study finds · arXiv research
A new study published on arXiv evaluates the effectiveness of code language models in detecting security patches, finding that even large models miss a significant portion of vulnerability fixes. The research highlights…
-
AI generates 54 false security bugs, infiltrating official databases
A security vulnerability, dubbed "Allucinate," has been discovered where AI-generated false bug reports have infiltrated official security databases. These fabricated entries, numbering 54, were reportedly created by an…
-
Samsung confirms 38 Android security fixes ahead of Google
Samsung has released its August 2026 Security Maintenance Release bulletin, detailing 38 Common Vulnerabilities and Exposures (CVEs) that Google provided for Android devices. This proactive disclosure by Samsung contras…
-
Microsoft Copilot faces security scrutiny over data leak risks · 4 sources tracked
Recent blog posts highlight significant security concerns surrounding Microsoft Copilot, suggesting the AI tool may pose data leak risks. Despite promises of enhanced productivity, findings like the 'Copirate 365' explo…
-
Google deploys AI agents to find and fix Chrome vulnerabilities at unprecedented speed
Google is leveraging AI, specifically Gemini agents, to significantly accelerate the process of finding, validating, and patching security vulnerabilities in Google Chrome. These AI systems can identify bugs, including …
-
AI generating fake vulnerabilities pollutes security pipeline
Artificial intelligence is now being used to generate fake vulnerabilities, polluting the Common Vulnerabilities and Exposures (CVE) pipeline. This trend is concerning security researchers who note that AI-generated vul…
-
Critical SQLite vulnerability patched after initial LLM hallucination confusion · 2 sources tracked
A critical vulnerability, CVE-2024-6377, has been identified in SQLite, a widely used database engine. This vulnerability was initially misattributed to an LLM's hallucination, but it is a genuine security flaw. The iss…
-
AI Agent Repositories Ranked by GitHub Stars, Usage, and Sentiment
A comprehensive analysis ranks 20 AI agent repositories based on various metrics including GitHub stars, daily token usage on OpenRouter, download counts from npm and PyPI, and historical vulnerability data. The ranking…
-
New AVE standard addresses AI agent behavioral vulnerability classification
A new open standard called AVE (Agentic Vulnerability Enumeration) has been developed to classify behavioral vulnerabilities in AI agents, addressing a gap left by existing systems like CVE and CWE. These traditional sy…
-
AI discovered 1,061 vulnerabilities, but exploitation rates are questioned · 1 source tracked
A recent report from VulnCheck analyzed vulnerabilities discovered or exploited in the first half of 2026. The analysis found 495 exploited vulnerabilities and identified 1,061 Common Vulnerabilities and Exposures (CVEs…
-
Mastodon poll asks users to predict CVE numbers for 2031 vs 2026
A poll on the social platform Mastodon asks users to predict whether the number of Common Vulnerabilities and Exposures (CVEs) will be higher or lower in 2031 compared to 2026. The poll, associated with the hashtag #Eva…
-
New paper maps CVEs to MITRE ATT&CK using expert-labeled data
A new research paper details a method for mapping Common Vulnerabilities and Exposures (CVEs) to MITRE ATT&CK techniques using a classifier trained on 1,207 expert-labeled CVEs. The study found that using LLM-generated …