Tego AI Reveals Second Claude Flaw: Files Sent to Attackers via Hidden Link

On July 24, 2026, Tego AI disclosed a second vulnerability in Anthropic's Claude ecosystem, following a previous report just a week earlier. This new flaw involves Claude Code, a command-line coding tool, which can be exploited to send files to attackers through a hidden link. The attack vector allows malicious actors to clone repositories and initiate unauthorized file transfers. The scope of the impact is significant, as it affects users of the Claude ecosystem, potentially compromising sensitive data. Tego AI's findings highlight the urgent need for security measures to mitigate these vulnerabilities. No specific CVEs were mentioned in the articles, and the current status of the vulnerabilities remains unclear, with no patches reported yet.

Key Points: • Tego AI disclosed a second vulnerability in Anthropic's Claude ecosystem within a week. • The new flaw allows attackers to send files via a hidden link in Claude Code. • No patches or CVEs have been released, indicating a potential ongoing risk.