Prompt and intellectual property protection.

[0wwafa]

Description of the feature request:

Some application "power" relies on the use of prompts.
Prompts unfortunately are unprotected.
No matter how you can obfuscate them, but they will always be visible in the debug console.

Solution:
publish a public key.
Allow the payload to be encrypted with that public key.

In that way, the application could encrypt the messages normally (or send them in plain text) but could pre-encrypt the prompts.

What problem are you trying to solve with this feature?

Intellectual property protection using APIs.

Any other information you'd like to share?

Well. to be hones this should be implemented on all APIs, but let's start from here.

[IvanLH]

Just to make sure I understand your ask correctly: You are looking to run this SDK on client devices, and want to be able to set up encrypted prompts, which will then be decrypted on by the Gemini API upon receipt?

[0wwafa]

Just to make sure I understand your ask correctly: You are looking to run this SDK on client devices, and want to be able to set up encrypted prompts, which will then be decrypted on by the Gemini API upon receipt?

Yes.