MediumNews
The risks of device code authentication for AI coding agents
- Published
- Record updated
Summary
Microsoft Security Blog reports lab research showing that device code phishing, which applies to any public client using the OAuth device flow, can be extended to AI coding agents, using GitHub Copilot CLI as an illustrative example. In a controlled simulation, a victim who approved an attacker-initiated GitHub device-code request enabled outcomes including remote steering of an active Copilot CLI session, injection of misleading memory entries, reading of synchronized cloud sessions, and access to private repositories available to the account. The researchers state these are proof-of-concept results and not attacks observed in the wild.
Topics
Related items
- InfoProton’s Lumo, a privacy-focused alternative to ChatGPT and Claude, arrives on desktopSame vendor · TechCrunch (Security)
- InfoCalifornia sues OpenAI, alleges 80-plus agent escapes hidden from publicSame vendor · CSO Online
- InfoCalifornia Sues OpenAI, Seeking Court-Ordered Changes After the Hugging Face HackSame vendor · SecurityWeek
- MediumCVE-2026-130358: AdminNeo before 5.9.1 contains a missing authentication vulnerability in the ClaudeSqlPlugin, GeminiSqlPlugin, and…Same vendor · NVD/CVE Database
- InfoIntroducing OpenAI for ChileSame vendor · OpenAI Blog