Chainlit AI Framework Flaws Enable Data Theft via File Read and SSRF Bugs

Summary: Zafran Security found two high-severity flaws in the open-source Chainlit AI framework, collectively called ChainLeak, that allow arbitrary file reads and server-side request forgery against servers hosting AI applications. The file-read bug (CVE-2026-22218) can expose /proc/self/environ for API keys and credentials, and the SSRF bug can reach the 169.254.169.254 metadata endpoint on AWS EC2 instances running IMDSv1; Chainlit fixed both in version 2.9.4 on December 24, 2025, after disclosure on November 23, 2025.
Bottom line: Chainlit has been downloaded 7.3 million times to date, and a single file read of /proc/self/environ hands an attacker the API keys and credentials needed to move laterally through your cloud environment.

Read the original at The Hacker News ↗

Topics: Vulnerabilities & Patching
Mentions: CVE-2026-22218
Related: ⚡ Weekly Recap: Firewall Flaws, AI-Built Malware, Browser Traps, Critical CVEs & More Cyber Risks ·
Attackers Exploit UnrealIRCd Flaw That Enables Server-Level Command Injection Cyber Risks · Nov 14, 2026
U.S. CISA adds Check Point and UnrealIRCd flaws to its Known Exploited Vulnerabilities catalog Cyber Risks · Nov 14, 2026