NewsDark Reading
Attackers Crash Claude's 'Office Party' Flaw to Run Malicious Code on Linux
Teaser from the source
Attackers can exploit a flaw in Claude's document-processing tool by creating a temporary file that Anthropic's AI assistant can access and run on a user's machine.
This is the RSS-feed teaser. Read the full article at the original source.
Read at Dark ReadingMore on Malicious
Other advisories
- MediumCVE-2026-121520Malicious client could set xattr value outside of BuildKit state directory
- Criticalghsa:GHSA-vghh-5rfx-xhq8LXD has arbitrary file read+write on host via rootfs/ symlink in malicious image
- Criticalghsa:GHSA-jpf8-86f3-wp38LXD has arbitrary file read+write on host via templates/ symlink in malicious image
- CriticalCVE-2026-48755LXD host escapes via malicious images, backup compression, and project-restriction bypasses (GHSA-jpf8 / GHSA-9j25 / GHSA-vghh / GHSA-pjff / GHSA-fmc8 / GHSA-47w9 / GHSA-7mr3 / GHSA-qx75)
- Highosv:MAL-2026-18657Malicious code in strands-aidr (PyPI)
- Highosv:MAL-2026-18651Malicious code in @secauditb20y/sec-test-b1 (npm)
- Highosv:MAL-2026-18650Malicious code in @secaudita10x/sec-test-a1 (npm)
- Highosv:MAL-2026-18652Malicious code in doghack-afterglow (npm)
Other news entries
- Hackers Compromise Ethersphere's bee-js and Swarm Desktop to Deliver Wallet-Stealing MalwareNewsThe Hacker News
- Tensorlake: attacker reached engineering GitHub via vendor-hosted sandboxes after unrotated env varsBreachpwn.watch
- Malicious Browser Extensions Could Use Claude Dispatch to Run Code on Users' ComputersNewsThe Hacker News
- CPython flaw exposes AI inference servers to sandbox escape attacksNewsCSO Online
- Email threat landscape: Q3 2026 trends and insightsNewsMicrosoft Threat Intelligence
- GhostMap Android flaw lets malicious app trigger persistent boot loopNewsBleepingComputer
- The Sandbox Fights Back: Joe Security Launches Adaptive DetonationNewsCSO Online
- Attackers turned OpsShield’s trusted update channel into a supply-chain attackNewsCSO Online
ID