← Back to the boardNews
Attackers Crash Claude's 'Office Party' Flaw to Run Malicious Code on Linux
Why it’s here: A specialist report on a vulnerability in Anthropic's Claude document-processing skills that enables code execution on enterprise Linux systems bears on AI and cybersecurity policy and enterprise AI agent governance, though it lacks an explicit federal policy hook.
Trace across the boardItem
6d agoDark Reading · research by AIR SecurityAttackers Crash Claude's 'Office Party' Flaw to Run Malicious Code on LinuxResearchers at AIR Security disclosed the "Office Party" flaw in Anthropic's Claude DOCX/PPTX/XLSX skills, which lets an attacker run code on Linux via a planted shared library that Anthropic has chosen not to fix.1 outletCyberAI37 linked
Connected
36 links- NewsMeasuring AI agent autonomy in practiceBoth involve Anthropic, Claude Code, Claude
- NewsWhen AI builds itselfBoth involve Anthropic, Claude Code, Claude
- NewsAnthropic Economic Index report: Learning curvesBoth involve Anthropic, Claude Code, Claude
- NewsDetecting and countering misuse of AI: August 2025Both involve Anthropic, Claude Code, Claude
- NewsEnabling independent research on how people use ClaudeBoth involve Anthropic, Claude Code, Claude
- NewsClaude's new constitutionBoth involve Anthropic, Claude
- NewsDetecting and countering malicious uses of ClaudeBoth involve Anthropic, Claude
- NewsHow Claude's text watermarking worksBoth involve Anthropic, Claude
- NewsAn update on our election safeguardsBoth involve Anthropic, Claude
- NewsWhat 81,000 people told us about AI economicsBoth involve Anthropic, Claude
- NewsDevs Hit by Trojanized Claude Code Extension for CursorBoth involve Anthropic, Claude Code
- NewsAnthropic Rolls Out Mythos 5.2 With Improved Bug Hunting, Fewer Sandbox EscapesBoth involve Anthropic, Claude
- NewsStatement on the comments from Secretary of War Pete HegsethBoth involve Anthropic, Claude
- NewsLLMs are making formal software verification feasible at a much larger scaleBoth involve Anthropic, Claude
- NewsMeasurements for understanding the pace of AI development inside frontier labsBoth involve Anthropic, Claude
- NewsImproving our alignment and security practicesBoth involve Anthropic, Claude
- NewsInvestigating three incidents in our cybersecurity evaluationsBoth involve Anthropic, Claude
- NewsU.S. elections readinessBoth involve Anthropic, Claude
- NewsClaude does cyber competitionsBoth involve Anthropic, Claude
- NewsMapping the mind of a large language modelBoth involve Anthropic, Claude
- NewsAutomated researchers can reliably mitigate alignment failuresBoth involve Anthropic, Claude
- NewsAnthropic targets valuation of up to $1.4 trillion for blockbuster IPOBoth involve Anthropic, Claude
- NewsInvestigating unintended model actions in our evaluations and internal useBoth involve Anthropic, Claude
- NewsLabor market impacts of AI: A new measureBoth involve Anthropic, Claude
- NewsOpenAI, Google, Anthropic hit with lawsuit seeking halt to frontier AI developmentBoth involve Anthropic, Claude
- NewsLLMs and bioriskBoth involve Anthropic, Claude
- NewsTeaching Claude whyBoth involve Anthropic, Claude
- NewsNatural Language AutoencodersBoth involve Anthropic, Claude
- NewsTrustworthy agents in practiceBoth involve Anthropic, Claude Code
- NewsCursor confirms security breach led to malicious Claude Code extensionBoth involve Anthropic, Claude Code
- NewsChallenges in red teaming AI systemsBoth involve Anthropic, Claude
- NewsEmotion concepts in a large language modelBoth involve Anthropic, Claude
- NewsReflections on our Responsible Scaling PolicyBoth involve Anthropic, Claude
- NewsAI Coding Agents Leak Thousands of Secrets to GitHubBoth involve Claude Code
- NewsHacking GitHub with a Semicolon & Claude with Sagi Tzadik | WizBoth involve Claude
- NewsMCPJacking: 155 Hijackable MCPs Discovered Live in the Official MCP MarketplaceBoth involve AIR Security
About
- Anthropic
- Claude Code
- Claude