THE DAILY EDITION

Cyber security

8 items · 60 articles in this topic

Exploitation of edge and management software dominated the day, with SAP, Barracuda, N-able and OpenSSH all in the frame. Supply-chain compromises of cloud credentials and package pipelines continued, and France formally blamed Iran for a physical-effect attack.

  1. 1

    Barracuda Email Security Gateway: two patched flaws are now exploited in the wild to run commands as root, per Barracuda's own warning.

    CriticalBleepingComputer ↗2 min read
  2. 2

    Google patched nine Chrome vulnerabilities, including CVE-2026-125350, a use-after-free exploited in the wild. It is the eighth actively exploited Chrome zero-day of 2026.

    HighThe Hacker News ↗3 min read
  3. 3

    Apple's November updates fix nine vulnerabilities, including CVE-2026-113971 affecting iOS 26 and Bluetooth pairing flaws. They also fix an iMessage group chat flaw that exposed messages to outside joiners.

  4. 4

    Qubole supply-chain attack: per Wiz, an attacker used Qubole's platform credentials to access multiple customers' AWS accounts from October 14 to 26 and steal secrets and data. Wiz also describes the broader pattern of backdoored trusted AWS integrations.

    HighThe Hacker News ↗6 min read
  5. 5

    Ethersphere's release pipeline was compromised to publish malicious bee-js 13.3.0 and Swarm Desktop 0.56.2, which steal wallet keys.

    CriticalThe Hacker News ↗8 min read
  6. 6

    France attributed an October sabotage attack on refrigeration controllers at about 700 retail sites to Iran's IRGC cyber unit. BleepingComputer reports fires and food losses, and TechCrunch says a second wave hit blood-bank freezers.

    HighThe Hacker News ↗6 min read
  7. 7

    ESET reports Turla is using a new modular malware, Cereal, to steal email from government Microsoft Exchange servers.

    HighThe Hacker News ↗5 min read
  8. 8

    Claude Code's 'Do not ask again' prompt is leading users to commit settings files that publish passwords and API keys to GitHub, per Ars Technica. The report cites thousands of exposed secrets.

    HighArs Technica ↗3 min read
More in Cyber security · 52 articles›
  • Binance warns that a public iOS sandbox escape named worse_query exposes wallet keys on iOS 26 and 27, with no Apple fix yet.

    The Hacker Newsiossandbox-escapecryptozero-day
    Feed excerpt

    An iOS sandbox escape published on GitHub lets an app read other apps' files on iPhones running iOS 26 and 27. That puts seed phrases, private keys, and login credentials at risk of exposure to malicious apps, Binance warned today. Apple's updates for iOS 26 and 27, released yesterday, do not list a fix. A researcher who goes by forcequit released the escape, called worse_query, on Sunday, a

  • SAP's November security updates fix 35 flaws, including four maximum-severity issues in NetWeaver, Web Dispatcher, and Gateway.

    BleepingComputersappatchescvenetweaver
    Feed excerpt

    SAP patched 35 vulnerabilities in its November 2026 security updates, including four maximum-severity issues affecting its NetWeaver, Web Dispatcher, and Gateway software. [...]

  • A decade-old OpenSSH flaw with musl, dubbed sshole, lets attackers log into Alpine and musl-based accounts using oversized passwords.

    Schneier on Securitycveopensshalpineauthentication-bypass
    Feed excerpt

    It’s a bad one, and it’s been in the code for a decade: A newly disclosed flaw involving OpenSSH and musl, dubbed “*sshole,” allows attackers to log in to certain Alpine Linux and other musl-based Linux accounts configured to block password authentication simply by entering an oversized password. “Oversized” means longer than 256 bytes. Details. Slashdot thread. And another news article.

  • N-able took N-central hosted environments offline after a researcher published 33 zero-days with PoC exploits, tracked as Able2Shell.

    The Hacker Newszero-dayn-ablepoc
    Feed excerpt

    N-able said it has taken N-central Hosted Environments offline and is urging on-premises customers to shut down their servers until further notice following the public disclosure of over 30 vulnerabilities collectively tracked as Able2Shell. "As a precautionary measure, we are temporarily taking our Hosted N-central environment offline while we take the necessary actions to secure the

  • N-able took its hosted N-central cloud offline after a researcher publicly disclosed over 30 alleged zero-day vulnerabilities, some enabling takeover.

    BleepingComputerzero-dayn-ablermmcloud
    Feed excerpt

    N-able has taken its hosted N-central environments offline and urged on-premises customers to shut down their servers after a security researcher publicly disclosed more than 30 alleged zero-day vulnerabilities, including exploits that could let attackers take over vulnerable systems. [...]

  • A vulnerability dubbed *sshole that lets attackers SSH into Linux devices is under active exploitation.

    Ars Technicavulnerabilitylinuxsshactively-exploited
    Feed excerpt

    "*sshole" poses a serious threat to Internet security.

  • Attackers are actively exploiting the SSHole OpenSSH vulnerability CVE-2026-123158 to run commands and steal credentials on Linux systems.

    BleepingComputercveopensshlinuxexploited
    Feed excerpt

    Hackers are actively exploiting the "SSHole" vulnerability (CVE-2026-123158) in OpenSSH to execute commands, steal credentials, and establish persistent access. [...]

  • Terminal2Shell is a critical flaw granting instant shell access across all major operating systems.

    Lobstersvulnerabilitycriticalrce
    Feed excerpt

    Comments

  • Google patched another actively exploited Chrome memory-corruption zero-day, bringing the 2026 count to eight.

    The Registerzero-daychromeexploitedbrowser
    Feed excerpt

    Another Chrome memory bug gets squashed, but not before attackers got their hands on it

  • Vanderbilt University systems were hit by a cyberattack with claimed extensive data theft, forcing restricted access and Wi-Fi changes.

    The Registerransomwareuniversitydata-breach
    Feed excerpt

    BYOD claims extensive data theft as campus Wi-Fi returns, most systems remain restricted, and classes carry on

  • Apple patched an iMessage group chat flaw exposing messages to outside joiners, plus Bluetooth pairing vulnerabilities.

    The Hacker Newsappleimessagepatch
    Feed excerpt

    An iMessage flaw could let a remote attacker join a group chat and receive its messages, Apple said yesterday, and it released updates to fix it. The updates also fix a separate iMessage flaw that could let someone added to a group access messages sent before they joined. As of today, other updates released yesterday for the iPhone XS, XS Max and XR, the iPad (7th generation) and some Apple

  • SAP's monthly Security Patch Day fixed four unauthenticated CVSS 10.0 flaws, including two enabling remote code execution in NetWeaver.

    The Hacker Newscvesaprcepatch
    Feed excerpt

    Four flaws that attackers could exploit without logging in were fixed today as part of SAP's monthly Security Patch Day—the vendor rates all four 10.0, the maximum CVSS score. Three are in the SAP NetWeaver application server, the software SAP systems run on, and one is in the Web Dispatcher, which receives web requests from the internet and passes them on to that server. The flaws in the Web

  • AI infrastructure startup Fireworks disclosed attackers accessed its development machines after breaching a third-party cloud provider.

    BleepingComputerbreachcloudsupply-chain
    Feed excerpt

    AI infrastructure startup Fireworks has disclosed a new security breach in which attackers accessed its development machines after breaching a third-party cloud provider. [...]

  • CISA added flaws in OpenSSH, GPUStack, FreeSWITCH and IBM Aspera Faspex to its KEV catalog for active exploitation.

    The Hacker Newscvekevactively-exploited
    Feed excerpt

    The U.S. Cybersecurity and Infrastructure Security Agency (CISA) on Monday added four security flaws impacting GPUStack, SignalWire FreeSwitch, OpenBSD OpenSSH, and IBM Aspera Faspex to its Known Exploited Vulnerabilities (KEV) catalog, citing evidence of active exploitation. The vulnerabilities are listed below - CVE-2026-123218 (CVSS score: 9.8) - An incorrect authorization vulnerability

  • Vanderbilt University confirmed a cyberattack by the BYOD ransomware gang that claims stolen student, employee, and research data.

    BleepingComputerransomwaredata-theftuniversity
    Feed excerpt

    Vanderbilt University warns it is responding to a cyberattack claimed by the BYOD ransomware gang, which claims to have encrypted the university's databases and stolen sensitive student, employee, and research data. [...]

  • Wyndham left a hardcoded credential in its customer app, exposing customer data to anyone who found it.

    TechCrunchhardcoded-credentialdata-exposurewyndham
    Feed excerpt

    Exclusive: Wyndham left a hardcoded credential in its customer app, allowing anyone to access customer data. The hotel giant wouldn't say if it has the ability to check for any past unauthorized access.

  • France attributes a sabotage-focused cyber campaign that disrupted French supermarkets to Iran.

    The Registeriransabotagenation-state
    Feed excerpt

    The 'first cyber offensive campaign carried out purely for the purpose of sabotage on national territory'

  • Sentra reports a supply chain attack compromising Qubole customers, with details published on its breaches blog.

    Hacker News (front page)supply-chainbreachcloud
    Feed excerpt

    Article URL: https://sentra.io/blog/breaches/inside-the-qubole-supply-chain-attack Comments URL: https://news.ycombinator.com/item?id=50432147 Points: 15 # Comments: 6

  • Wiz researchers uncovered a supply-chain attack that backdoors trusted AWS integrations to steal secrets.

    The Registerawssupply-chainsecretscloud
    Feed excerpt

    Wiz researchers uncovered the secret-stealing supply-chain attack

  • France's foreign minister blamed Iran for two waves of cyberattacks that hit industrial freezers in grocery stores and blood banks.

    TechCrunchcyberattackirancritical-infrastructure
    Feed excerpt

    France's foreign minister blamed Iran for launching at least two waves of cyberattacks targeting industrial refrigerated and freezer units across France and wider Europe.

  • France attributed October cyberattacks that disrupted refrigeration at over 700 supermarkets, causing fires and food losses, to Iranian state-backed hackers.

    BleepingComputerattributionnation-stateiranfranceics
    Feed excerpt

    France has linked October cyberattacks that disrupted refrigeration systems at more than 700 supermarkets and other retail locations to Iranian state-backed hackers, saying the attacks caused food losses and sparked at least 15 fires. [...]

  • Russian cloud provider Beget is rebuilding its cloud infrastructure from scratch after a data-wiping malware attack.

    BleepingComputerwipercloudransomware
    Feed excerpt

    A data-wiping malware attack has forced Russian cloud services provider Beget to rebuild its cloud infrastructure from scratch, the company said. [...]

  • Researchers chained WAF bypasses and default logins in automakers' Android apps to remotely unlock cars and honk horns in Australia and the US.

    The Registervehicle-securityandroidvulnerability
    Feed excerpt

    Pair of researchers chain WAF bypasses and default logins to command vehicles in Australia and the US

  • Fortinet reports the Bitter APT group is using a new C++/Rust remote access trojan, PapRAT, against South Asian targets.

    The Hacker Newsaptmalwareespionagerat
    Feed excerpt

    The South Asian cyber espionage group known as Bitter (aka APT-C-08, APT-Q-37, Hazy Tiger, Orange Yali, T-APT-17, and TA397) has been observed targeting South Asian organizations with a previously unreported remote access trojan called PapRAT. "PapRAT has both C++ and Rust implementations," Fortinet FortiGuard Labs researcher Cara Lin said. "Its capabilities include command execution, file

  • Eight additional X.Org Server vulnerabilities were disclosed, also affecting XWayland, with the reports reportedly AI/LLM-assisted.

    Phoronixx.orgxwaylandvulnerabilityopen-source
    Feed excerpt

    With the assistance of AI/LLMs, eight more security vulnerabilities were made public today for the X.Org Server. These issues also end up affecting the XWayland code too...

  • Queens Public Library suspended e-borrowing services after a ransomware attack on its systems.

    Hacker News (front page)ransomwarebreach
    Feed excerpt

    Article URL: https://connect.queenslibrary.org/ Comments URL: https://news.ycombinator.com/item?id=50437790 Points: 58 # Comments: 37

  • Introduces meta-benchmarking that converts existing LLM agent utility benchmarks into indirect prompt injection evaluation suites.

    arXiv cs.CR (Cryptography & Security)prompt-injectionllmbenchmark
    Feed excerpt

    arXiv:2611.10026v1 Announce Type: new Abstract: Large language model (LLM) agents increasingly perform complex tasks using high-privilege tools, making indirect prompt injection (IPI) a critical security risk. However, IPI benchmarks lag behind their rapidly evolving utility counterparts, only offering limited task and environment coverage and relying on narrow, static attack scenarios. In this paper, we propose meta-benchmarking for IPI: transforming existing utility benchmarks into end-to-end IPI evaluation suites that reuse their realistic tasks, environments, and infrastructure. We present MetaPI, an extensible toolkit that automatically identifies plausible injection points, generates realistic attack objectives, optimizes adaptive payloads, and validates attacks in the real environment. Its modular interfaces support heterogeneous environments, agents, defenses, and attack algorithms, allowing new components to be integrated without redesigning the evaluation pipeline. We apply MetaPI to AppWorld, VitaBench, and Terminal-Bench and evaluate representative agents and defenses across multiple backbone LLMs. Our results show that no evaluated defense simultaneously achieves stron…

  • Introduces SEA, an interactive benchmark evaluating personal AI agents' resistance to social engineering attacks across 320 scenarios.

    arXiv cs.CR (Cryptography & Security)prompt-injectionagentsbenchmark
    Feed excerpt

    arXiv:2611.10037v1 Announce Type: new Abstract: Personal AI agents are increasingly deployed on user devices to execute actions involving real money and sensitive data, but their security under social engineering attacks has not been systematically evaluated. We propose SEA (Social Engineering Attack Benchmark), the first interactive environment for evaluating the intrinsic security capability of AI agents in social engineering attack scenarios, covering 8 daily-life domains, 30+ simulated applications, a 4-level difficulty gradient, and a 5-dimensional evaluation framework, with 320 evaluation tasks designed in consultation with anti-fraud specialists and spanning two culturally distinct settings in Chinese and English. Evaluation on 11 frontier open-source and closed-source models, with GPT-6 evaluated only in Chinese, reveals that all models are breached to varying degrees (Chinese attack success rates ranging from 16.9% to 84.3%) proactive verification rates universally below 32%. We further propose META-SEA, a co-evolutionary framework for attack tasks and defenders. With META-SEA training, Qwen3-8B achieves defense performance comparable to closed-source models while effectiv…

  • Presents BACGUARD, a reverse proxy that enforces authorization on unmodified applications to mitigate broken access control.

    arXiv cs.CR (Cryptography & Security)access-controlowaspweb
    Feed excerpt

    arXiv:2611.10157v1 Announce Type: new Abstract: Broken Access Control (BAC) has been the foremost category in the OWASP Top Ten since 2021, primarily due to the challenges in enforcing authorization at the application layer. Traditional mechanisms depend on logic embedded in the application, referencing relational database information such as permissions, roles, or ownership. However, missing or incorrect checks can go undetected by external systems, leaving applications vulnerable. This paper presents BACGUARD, a reverse proxy designed to enforce authorization for unmodified applications. BACGUARD reconstructs authorization relations by leveraging authenticated user data from session stores and object ownership from relational schemas. It evaluates whether users are permitted to access resources and blocks unauthorized requests in real time. Unlike existing approaches, BACGUARD bases its authorization on database joins, making the system resilient to forgery. For most requests, decisions are made deterministically. In cases where access control violations are only apparent in response data, such as cross-user data leakage, a language model is used as a fallback. Evaluations on six…

  • Proposes task-level backdoor attacks that poison self-evolving LLM agents by implanting malicious rules into their distilled skills.

    arXiv cs.CR (Cryptography & Security)llm-agentsbackdoorprompt-injection
    Feed excerpt

    arXiv:2611.10542v1 Announce Type: new Abstract: Large language model (LLM) agents increasingly improve themselves through self-evolution, distilling execution experience into persistent, reusable skills. However, this persistence creates a new attack surface: malicious rules implanted into skills can compromise subsequent agent behavior. Existing skill backdoor attacks typically assume direct manipulation of the victim's skills, which is impractical when skills are constructed internally by a trusted evolver. In this paper, we study a novel problem of task-level backdoor attacks, where the attacker can only inject poisoned evolution tasks, inducing the agent to implant backdoors into its own skills. Since the attacker controls neither skill induction nor downstream execution, a natural approach is to iteratively optimize poisoned tasks based on attack success. However, our preliminary analysis reveals two key challenges: (i) failures at different stages of skill evolution and downstream execution can produce indistinguishable outcomes, providing little guidance for targeted poisoned-task updates; and (ii) unsuccessful updates may undermine earlier progress, while simply discarding …

  • Analyzes 25 confirmed cyberattacks on German-speaking universities from 2022-2023 and documents a phishing cluster's TTPs.

    arXiv cs.CR (Cryptography & Security)phishingransomwareeducation
    Feed excerpt

    arXiv:2611.10698v1 Announce Type: new Abstract: In light of recent ransomware attacks, this study analyzes and measures the impact of cyberattacks against academic institutions in German-speaking areas. First, we provide an overview of 25 confirmed cyberattacks against universities in German-speaking areas between February 2022 and February 2023. In this time frame, we find that 5% of all German universities have experienced a confirmed cyberattack with a measurable impact. Second, we document victimology and Tactics, Techniques, and Procedures (TTP) of a phishing activity cluster spanning more than three years and showing a particular focus on universities in German-speaking areas. Of the 93 targeted universities, 58 (62%) are German, representing a significant fraction with approximately 14% of the total number of universities (423) in Germany. A notable technique is that the phishing emails are sent via university infrastructure, using valid, compromised credentials of university-associated personnel and involving university Virtual Private Network (VPN) tunnels.

  • Proposes topology-aware isolation as a provable defense against timing side channels in shared LLM KV caches.

    arXiv cs.CR (Cryptography & Security)llmside-channelmulti-tenant
    Feed excerpt

    arXiv:2611.10756v1 Announce Type: new Abstract: Large Language Models (LLMs) are increasingly deployed as shared multi-tenant services, relying on advanced optimizations like KV cache and prefix sharing to improve throughput and reduce latency. However, these optimizations introduce a critical vulnerability: attackers can exploit time-to-first-token (TTFT) variations between KV cache hits and misses to infer sensitive prompt tokens via prefix-matching side-channel attacks. Existing defenses indiscriminately restrict cross-tenant cache sharing, which incurs prohibitive performance overhead and negates the fundamental efficiency benefits of the KV cache. We propose Mirage, a topology-aware emission obfuscation framework that provably bounds timing leakage without sacrificing the performance gains of cache sharing. Instead of stalling the GPU or forcing redundant recomputations, Mirage delays token emission at the gateway so the attacker observes a natural cache miss distribution, while the GPU seamlessly continues serving other requests at full speed. Mirage uses a Privacy Odometer that persistently tracks the topological novelty of prefix fan-out expansions to distinguish aggressive…

  • Evaluates three open-source scanners on 140 agent skills to detect malicious instructions hidden in AI agent skill files.

    arXiv cs.CR (Cryptography & Security)ai-agentssupply-chainmalware
    Feed excerpt

    arXiv:2611.10830v1 Announce Type: new Abstract: Agentic AI ecosystems increasingly rely on reusable agent skills that combine executable code, natural-language instructions, configuration artifacts, and external tool integrations. These hybrid artifacts create software supply-chain risks because malicious intent may be distributed across instructions, code, metadata, and runtime behavior, making isolated alerts difficult to interpret. We evaluate three open-source scanners on a held-out test set of 140 skills comprising 80 malicious and 60 benign samples, including Cisco Skill Scanner with and without its optional LLM analyzer. The default Cisco configuration achieves a recall of 0.91 with 33 false positives among the 60 benign skills, while the LLM-enabled configuration using Qwen3:8B achieves a recall of 0.85 with 40 false positives. We then present a framework that combines static findings, safety-constrained sandbox evidence, and LLM-based reasoning to assess malicious intent. The framework tracks the origins of findings and relates evidence across analysis layers. In a within-system evidence-source ablation using the same reasoning backend, adding dynamic evidence to skill-con…

  • Systematizes known malicious attacks on agentic AI systems, covering tools, memory, skills, and multi-agent coordination risks.

    arXiv cs.CR (Cryptography & Security)ai-agentssokprompt-injection
    Feed excerpt

    arXiv:2611.10843v1 Announce Type: new Abstract: Agentic AI systems extend large language models with tools, memory, retrieval, browser or computer-use interfaces, code execution, external services, and multi-agent coordination. Although these capabilities enable agents to act beyond text generation and perform tasks, they also create attack paths through which adversaries can exploit untrusted content, deceptive metadata, poisoned state, compromised skills, mutable dependencies, or malicious code. Such attacks can manipulate planning and tool use, corrupt memory, abuse code-execution environments, and trigger harmful downstream effects. In this paper, we present a systematic review of malicious attacks targeting agentic AI systems. We screened 186 candidate sources and retained 98. Following independent coding and resolver adjudication, we identified 284 atomic attacks, 452 atomic safeguards, and 660 explicit attack--safeguard relationships. We organize attacks into eight causal-mechanism classes while retaining insertion points, activation conditions, abused capabilities, objectives, impacts, persistence, and evidence as orthogonal dimensions. Our analysis shows that tool, MCP, an…

  • Validates 233 AI-discovered cloud threats on live AWS, GCP, and Azure infrastructure, confirming 103 with 75.2% precision.

    arXiv cs.CR (Cryptography & Security)cloudthreat-modelingllm
    Feed excerpt

    arXiv:2611.10947v1 Announce Type: new Abstract: Cloud infrastructure threat modeling lacks empirical validation of AI-discovered threats. We present a three-phase methodology of multi-agent threat discovery, automated exploit synthesis, and non-destructive live validation, and apply it to three commercial foundation models on two cloud architectures, validating each threat natively on the cloud (AWS, GCP, or Azure) it was discovered against. Of 233 AI-discovered threats, 103 were confirmed and 34 refuted (75.2\% precision, 95\% CI [67.3, 81.7]); 44 were conditional (the named mechanism refuted but an equivalent impact shown through another path) and 52 indeterminate. Multi-model discovery helped through breadth, not agreement: the three models together found 2.2$\times$ the confirmed threats of the best single model at comparable precision, while threats reported by several models were confirmed no more often than single-model ones. False positives were not architecture-specific: they asserted access or system state the deployment did not have. Threats coded in advance as observable configuration resolved cleanly more often than multi-step behavioral threats (72.5\% vs.\ 51.6\%), w…

  • Argues that jailbreak detector evaluation should focus on low false-positive operating points rather than AUROC alone.

    arXiv cs.CR (Cryptography & Security)jailbreakllmevaluation
    Feed excerpt

    arXiv:2611.11067v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly deployed in user-facing systems, where jailbreak attacks can bypass safety controls and induce harmful and policy-violating outputs. Recent model-internal detectors use hidden representations, gradients, or internal signals to identify such prompts, but their practical security value depends critically on behavior at low false-positive operating points rather than on area under the receiver operating characteristic curve (AUROC) alone. We study two data dependencies that shape this behavior: calibration data, which determine the detection threshold, and reference data, which can influence detector scores. Across four model-internal detectors, four LLMs, and twelve jailbreak families, thresholds calibrated on curated benign prompts frequently fail to preserve their false-positive rate (FPR) budgets on real-user traffic. Recalibration improves false-positive control but reveals substantial family-level coverage gaps, including cases with high AUROC but low true-positive rate (TPR). Causal reference-set interventions show that changing reference composition can substantially alter family-leve…

  • Researchers build a curated dataset of MCP security vulnerabilities drawn from CVE disclosures to enable systematic study.

    arXiv cs.CR (Cryptography & Security)mcpcvedatasetllm-agents
    Feed excerpt

    arXiv:2611.11104v1 Announce Type: new Abstract: As a core protocol for integrating large language models (LLMs) with external data sources and tools, the Model Context Protocol (MCP) has quickly become a key interface in emerging AI agent systems. As the MCP ecosystem expands, reported vulnerabilities have also increased, creating a growing software-security attack surface. However, MCP vulnerability information remains scattered, noisy, and difficult to study systematically. This paper addresses that gap by constructing a curated MCP vulnerability dataset from CVE disclosures. Starting from 119 MCP-related records retrieved through February 2026, we filter unrelated cases out and obtain a validated snapshot of 24 MCP vulnerabilities. Our filtering methodology explicitly addresses acronym ambiguity and achieves 96.7% inter-rater agreement on a sampled exclusion audit while retaining all known high-severity MCP cases identified in public advisories. Using this initial dataset snapshot, we find that 87.5% of validated vulnerabilities are Critical or High severity, and that injection and path traversal vulnerabilities account for 66.7% of cases. We further identify recurring issues in…

  • Proposes a runtime audit that checks LLM agent tool calls against an intent graph to block indirect prompt injection.

    arXiv cs.CR (Cryptography & Security)prompt-injectionllm-agentsleast-privilege
    Feed excerpt

    arXiv:2611.11305v1 Announce Type: new Abstract: LLM agents that call tools are exposed to indirect prompt injection: instructions hidden in tool outputs can steer an agent into calls the user never asked for. We present a runtime audit that derives an executable intent graph from the user's objective and the deployment's tool catalog with one LLM call per task, and checks every tool call against it with a deterministic loop that makes no LLM call (median 5.7 microseconds per call). Two further mechanisms recover utility. Lookups of the user's own state bypass the audit, because the harm of an attack lies in its write or outward step. The calls that remain flagged are escalated to a reasoning guardian, in the spirit of tool-integrated reasoning models such as START: a reasoning model that sees only the objective, the plan, the call history without tool outputs, and the flagged call, and can verify facts with a restricted Python-expression tool. A current agent showed no baseline attack success, so end-to-end validation remains open.

  • Study finds coding agents replicate vulnerabilities present in existing codebases when generating new code.

    arXiv cs.CR (Cryptography & Security)coding-agentsvulnerabilitiessecure-coding
    Feed excerpt

    arXiv:2611.11455v1 Announce Type: new Abstract: LLM-based coding agents are widely used to develop critical production software. Even though coding agents are known to occasionally write vulnerable code, there has not been any investigation into how code base vulnerabilities evolve over time under coding agent contributions. In this work, we close this gap by investigating whether and how vulnerabilities in existing code lead coding agents to introduce similar vulnerabilities in the code they write. We first evaluate coding agents on 96 tasks that require implementing a new backend application in a repository where other applications already exist. Second, we investigate repository-level code generation on 20 pairs of tasks in real C/C++ projects. In each task, the agent is asked to complete code critical to memory safety. We find that in both settings, insecure existing code leads to agents writing vulnerable code more often. In backend generation, insecure neighbors reduce the share of correct and secure implementations by 3.4% on average, and in C/C++ repository-level coding, the average gap between secure and insecure contexts reaches 35.4%. Prompting to follow existing code-st…

  • EVAX proposes adaptive defenses that activate only on detected microarchitectural attacks, cutting overhead by over 95%.

    arXiv cs.CR (Cryptography & Security)side-channelspeculative-executionhardware-security
    Feed excerpt

    arXiv:2611.11498v1 Announce Type: new Abstract: This paper provides an end-to-end solution to defend against known microarchitectural attacks such as speculative execution attacks, fault-injection attacks, covert and side channel attacks, and unknown or evasive versions of these attacks. Current defenses are attack specific and can have unacceptably high performance overhead. We propose an approach that reduces the overhead of state-of-art defenses by over 95%, by applying defenses only when attacks are detected. Many current proposed mitigations are not practical for deployment; for example, InvisiSpec has 27% overhead and Fencing has 74% overhead while protecting against only Spectre attacks. Other mitigations carry similar performance penalties. We reduce the overhead for InvisiSpec to 1.26% and for Fencing to 3.45% offering performance and security for not only spectre attacks but other known transient attacks as well, including the dangerous class of LVI and Rowhammer attacks, as well as covering a large set of future evasive and zero-day attacks. Critical to our approach is an accurate detector that is not fooled by evasive attacks and that can generalize to novel zero-day at…

  • Compares prompt-based and deterministic defences against prompt injection in tool-using agents, including open-weights model validation.

    arXiv cs.CR (Cryptography & Security)prompt-injectionllm-agentsreference-monitor
    Feed excerpt

    arXiv:2611.11571v1 Announce Type: new Abstract: Tool-using LLM agents act on instructions embedded in the data they read, so prompt injection is a structural rather than a prompt-wording problem. We compare in-prompt defences (hardened prompts, spotlighting, LLM detectors) with deterministic enforcement outside the model (four reference monitors of increasing strength), human approval and taint tracking, judged by one effect-based oracle independent of every defence in an executable environment of 13 tasks and 12 tools, with argument-fuzzing attackers and severity-weighted harm. In an evaluation with real open-weights models (Qwen2.5-1.5B and Qwen2.5-7B; 156 attack episodes per defence and model), the models proposed the attacker's call in 64% and 84% of attack episodes behind the task-bound monitor, which admitted 1.3% and 7.7%; guardrail effects depended on the model (a prompt guardrail moved obedience from 67% to 65% and from 86% to 55%, spotlighting plus a detector to 12% and 0%) against static attacks. Simulation with an assumed model covers what no real run did: under the assumed erosion law a persistent adaptive attacker or a compromised model drives every guardrail stack to…

  • Introduces ShadowBench-MCP, a benchmark measuring how LLM routers can be steered toward riskier tool execution paths without malicious requests.

    arXiv cs.CR (Cryptography & Security)llm-agentsbenchmarktool-routing
    Feed excerpt

    arXiv:2611.11720v1 Announce Type: new Abstract: LLM agents increasingly act as routers for tools, APIs, and software-supply-chain workflows, making an execution path - rather than generated text alone - a security-relevant model output. We study shadow invocation: a task-preserving request variant that causes an LLM router to select a riskier execution path, potentially skipping validation, provenance, staging, redaction, or audit controls even though the request does not ask for a malicious action. We introduce ShadowBench-MCP, a benchmark and metric suite that separates task preservation (TPSS), operational pressure (OPSS), and route-geometric drift (RGAS), and combines them in the pre-execution SODS score. In a focused policy-ablation study using a Groq-hosted qwen/qwen3-32b router, 120 routing decisions yield 96 original-versus-variant comparisons across six security-sensitive workflows. Efficiency-first routing produces a 29.2% shadow-invocation rate and 26.7% safety-tool skip rate. Policy-aware routing reduces these to 4.2% and 1.7%, respectively, corresponding to an 85.7% relative reduction in shadow invocations. Pre-execution signals distinguish shadow routes with ROC AUC u…

  • Uses agent swarms to scale adaptive prompt injection attacks against AI agents, arguing defenses remain necessary despite apparent saturation.

    arXiv cs.CR (Cryptography & Security)prompt-injectionai-agentsred-teaming
    Feed excerpt

    arXiv:2611.11861v1 Announce Type: new Abstract: Prompt injections have emerged as one of the top security risks faced by AI agents today. Early work in prompt injection attacks looked at simple tool-call agents and static attack payloads, and were saturated since mid-2025. As of Fall 2026, human red-teaming may be nearing saturation, especially on the Gemini family of models. In this work, we argue that despite this apparent plateau, prompt injection attacks are still possible and continued effort into prompt injection defense is needed. We design a new synthetic evaluation dataset which uses random non-malicious actions as adversarial goals. Such dataset design helps us to evaluate prompt injection robustness as a model ability to respect instruction hierarchy, and disentangle this ability from model knowledge of security practices. We show that an agentic attacker of increased capabilities can reliably trick model to follow instructions in tool output in our evaluation, even for models previously considered robust on public prompt injection benchmarks such as GraySwan Arena. This shows that current frontier defenses may require proactive, automated red-teaming to remain resilient…

  • Measures how execution harness design, not just the backend model, shapes the security resilience of web agents.

    arXiv cs.CR (Cryptography & Security)web-agentsllmharness
    Feed excerpt

    arXiv:2611.11971v1 Announce Type: new Abstract: Web agents extend LLMs from text generation to real-world web interaction by relying on execution harnesses that interpret user goals, perceive webpages, plan actions, and ground model decisions into browser operations. While prior work has recognized that agent security depends on more than the backend model alone, how harness design quantitatively shapes web-agent security remains insufficiently measured. To systematically study web-agent harness security, we first characterize web-agent harnesses as a four-stage execution workflow covering goal interpretation, webpage perception, reasoning and planning, and action execution. Based on this abstraction, we build WebTrap Farm, a systematic and extensible evaluation framework covering three representative attack categories: malicious user prompts, malicious prompt injection, and deceptive web design. To avoid conflating security with task capability, we use paired usability calibration tasks to separate genuine defensive behavior from capability limitations. Our evaluation shows that existing web-agent harnesses remain vulnerable to typical attacks, with some unsafe behaviors occurring…

  • Describes StealthyPoison, a procedural poisoning attack that corrupts decision rules in third-party skills used by LLM agents while preserving normal behavior.

    arXiv cs.CR (Cryptography & Security)data-poisoningllm-agentssupply-chain
    Feed excerpt

    arXiv:2611.12155v1 Announce Type: new Abstract: LLM agents increasingly retrieve third-party skills from public registries and follow their procedures to complete user tasks. This reliance allows attackers to manipulate agent behavior by poisoning the skills. However, overtly malicious instructions can provide explicit signals for detection. We present StealthyPoison, a stealthy procedural poisoning attack that corrupts a skill's decision-critical rules to induce incorrect task outcomes while preserving its advertised functionality and normal execution. Unlike prior knowledge corruption attacks that craft poisoned content for known target question-answer pairs, StealthyPoison constructs reusable poisoned procedures from a publicly available third-party skill, without access to actual user requests or their ground-truth answers. To make the poisoned procedure generalize to unseen requests, it first synthesizes requests from the target skill and executes its clean version using a surrogate agent to obtain pseudo-references. These references and surrogate execution feedback guide the iterative refinement of payloads that alter decision-critical rules. To increase the likelihood that t…

  • Proposes SNAPRAG, a query-efficient attack that extracts private knowledge bases from RAG systems using adversarial perturbations.

    arXiv cs.CR (Cryptography & Security)ragdata-extractionllm
    Feed excerpt

    arXiv:2611.12241v1 Announce Type: new Abstract: Retrieval-Augmented Generation (RAG) systems improve Large Language Model (LLM) responses by incorporating information from external knowledge bases. However, recent studies have shown that attackers can extract substantial portions of private knowledge bases from RAG systems through repeated querying. Yet, state-of-the-art extraction attacks either rely heavily on external LLMs to generate adversarial queries or require a large number of queries. In this paper, we propose SNAPRAG, a novel and cost-efficient attack for extracting RAG knowledge bases through adversarial perturbations. Our approach algorithmically generates adversarial queries without relying on external LLMs and allows for a strictly configurable query budget. In particular, SNAPRAG uses a discrete token-mutation algorithm to generate novel queries in a strict black-box setting. Our adaptive dissimilarity threshold calibration systematically forces queries into unexplored semantic regions of the target database, enabling extraction with fewer queries. Specifically, to extract half of the target knowledge base, our method requires up to 6.77$\times$ fewer queries than e…

  • Analyzes 2021-2026 cross-chain bridge hacks, finding 88% of losses since 2023 stem from security posture failures and proposing an assessment rubric.

    arXiv cs.CR (Cryptography & Security)cross-chainbridge-hackscrypto
    Feed excerpt

    arXiv:2611.12252v1 Announce Type: new Abstract: Bridge hacks have cost billions, often without exploiting a code bug. In the cross-chain bridge hacks we catalogue from 2021 to 2026, failures of security posture, meaning how a bridge's people, code, controls and off-chain infrastructure are set up, account for 88% of reported losses since 2023, and nearly all repeat earlier failures. Code audits cannot catch these failures, so we propose a bridge security assessment rubric. It scores each bridge deployment from on-chain evidence, and a hard cap drops the score to the lowest value when a fatal risk is present. A token is only as safe as its weakest deployment if all its deployments share one backing. For attacks nobody anticipated, we show that rate limits and an emergency stop still bound the loss. We build an analyzer that applies the rubric to deployed bridges and monitors them for changes, and an agentic analyzer that runs the same checks on any bridge. Operators learn what to fix, and users can compare bridges. Our rubric's controls cut bridge hack losses 105-fold, from \$3.52B to \$33M.

  • A project article examines flawed public key and certificate practices, illustrating bad certificate handling tricks.

    Hacker News (front page)tlscertificatescrypto
    Feed excerpt

    Article URL: https://maurycyz.com/projects/badcert/ Comments URL: https://news.ycombinator.com/item?id=50434264 Points: 45 # Comments: 18

  • SEGA disclosed a breach at newsletter provider ARARA that exposed email addresses of 2.2 million subscribers.

    BleepingComputerbreachsegaemail
    Feed excerpt

    Japanese gaming giant SEGA has disclosed that a data breach at its third-party newsletter provider, ARARA, exposed the email addresses of 2.2 million newsletter subscribers. [...]

  • York University disclosed a cybersecurity incident where a threat actor stole personal information submitted through online forms dating back to 2007.

    BleepingComputerbreachdata-theftuniversity
    Feed excerpt

    York University has announced it suffered a cybersecurity incident in which a threat actor stole personal information people submitted through online forms. [...]

  • SANS ISC reports attackers scanning honeypots for SAP NetWeaver webshells, using a newly observed URL pattern.

    SANS Internet Storm Centerwebshellsapscanning
    Feed excerpt

    Just saw a new (to me) URL in our honeypot data today:

  • Binance denied a data breach claimed by a threat actor using the ShinyHunters handle on an underground forum.

    BleepingComputerbreach-claimcryptoshinyhunters
    Feed excerpt

    Binance, the world's largest crypto exchange, has denied claims of a new security breach made by a threat actor using the 'ShinyHunters' handle on a newly emerged underground cybercrime forum. [...]

Written by claude-sonnet-5-5 from feed titles and summaries. The model can misstate details; every claim links to its source.