🛡️ Sentinel: [CRITICAL] Fix command injection in text-to-speech

[omsingh02]

🚨 Severity: CRITICAL

💡 Vulnerability: The FeedbackManager.speak_text function on Windows interpolated raw user input (text) directly into a string executed by subprocess.Popen(["powershell", "-Command", ...]). An attacker could exploit this by providing a maliciously crafted text string to break out of the string context and execute arbitrary PowerShell commands on the host machine.

🎯 Impact: Remote Code Execution (RCE) or local privilege escalation, depending on where the text input originates from within the application (e.g. metadata of a recognized song). An attacker could gain full control over the user's system running the application on Windows.

🔧 Fix:

• Replaced the direct string interpolation with a secure Base64 payload delivery mechanism.
• The Python code now Base64-encodes the text using utf-16le encoding.
• The PowerShell script decodes the Base64 string back into Unicode and passes it directly to $synth.Speak().
• This ensures that no special characters in the text can be interpreted as PowerShell syntax, completely mitigating the command injection risk.

✅ Verification:

• The Python compilation and test suite (pytest src/) execute without errors.
• Verified that the FeedbackManager.speak_text function continues to behave normally on all supported operating systems.

---

PR created automatically by Jules for task 18274673868750369412 started by @omsingh02

[google-labs-jules]

👋 Jules, reporting for duty! I'm here to lend a hand with this pull request.

When you start a review, I'll add a 👀 emoji to each comment to let you know I've read it. I'll focus on feedback directed at me and will do my best to stay out of conversations between you and other bots or reviewers to keep the noise down.

I'll push a commit with your requested changes shortly after. Please note there might be a delay between these steps, but rest assured I'm on the job!

For more direct control, you can switch me to Reactive Mode. When this mode is on, I will only act on comments where you specifically mention me with @jules. You can find this option in the Pull Request section of your global Jules UI settings. You can always switch back!

New to Jules? Learn more at jules.google/docs.

---

For security, I will only act on instructions from the user who triggered this task.