Papers & Research

PromptArmor discloses Databricks Genie malicious Skill that bypasses multiple controls for phishing and data exfiltration

Original title:Phishing in Databricks Genie

AI summary

PromptArmor disclosed that Databricks Genie Code executes a malicious Skill, displaying a phishing pop-up when users view analysis results and exfiltrating tenant data through the browser, all without human approval. Organization-level Skill governance, guardrail agents, coding-environment egress controls, and chat display sandboxes all failed to block the attack. After receiving the disclosure on August 16, 2026, Databricks determined that ensuring uploaded Skills contain no malicious content is the user's responsibility.

Key Takeaways

  • PromptArmor disclosed that Databricks Genie Code executes a malicious Skill,displaying a phishing…
  • Organization-level Skill governance,guardrail agents,coding-environment egress controls,and chat…
  • After receiving the disclosure on August 16,2026,Databricks determined that ensuring uploaded…

Quick Facts

Published
Updated
Source
PromptArmor:Threat Intelligence
Category
Papers & Research
AI score
64
Read the original Read the full article at the original source
Why it matters

报告指出组织依赖的多层控制均无法阻止恶意 Skill 的攻击路径,并附 Databricks 的回应,可供依赖 Skill 生态的团队核对自身防护假设。

Source & copyright

This article is based on curated titles and AI summaries from the AIHOT public API, with attribution and links back. Original copyright remains with the original publishers and authors; this site does not mirror full texts.

Original article · AIHOT entry

How to Cite

When citing this article, keep the title, date, and permalink, and credit Orange News as the source.

Orange News: PromptArmor discloses Databricks Genie malicious Skill that bypasses multiple controls for phishing and data exfiltration. October 5, 2026. https://new.nat6.net/en/news/dfcis3hxowljt45i0bvyi3f0n/

View all →