Cline Bot AI Coding Agent Data Exfiltration via Prompt Injection and DNS

Affected Vendor(s)

Affected Product(s)

Summary

Through malicious instructions planted in a source code file, Cline can be coerced into exfiltrating sensitive key material from a user’s environment to an attacker-controlled location.

Cline is vulnerable to prompt injection when analyzing source code files. Furthermore, this prompt injection can be used to execute what is considered a safe command (ping), which requires no user approval, in a way that will exfiltrate sensitive key material to an attacker-controlled location.

Timeline

24th August, 2025 Identified flaws
27th August 2025 Reported to vendor
31st October 2025 Product re-tested
18th November 2025 Findings published

Credit

Blog Post

References

Learn how Mindgard can help you navigate AI Security

Take the first step towards securing your AI. Book a demo now and we'll reach out to you.