Triggering a security action based on an AI-generated code package recommendation

A method implemented by a computing system, the method comprising:

• causing an artificial intelligence (AI) model to recommend an identified code package to resolve a coding problem by providing an AI prompt to the AI model,
• wherein the AI prompt requests that the AI model identify a code package that is written in a programming language and that comprises a designated functionality that resolves the coding problem;
• receiving a response to the AI prompt from the AI model, the response comprising a package identifier that identifies the identified code package;
• in response to receipt of the response, which comprises the package identifier that identifies the identified code package, from the AI model, analyzing a plurality of code packages for existence of the identified code package;
• based at least on confirmation of the existence of the identified code package, determining whether a value of an attribute of the identified code package satisfies a criterion associated with non-trustworthiness;
• and
• based at least on the value of the attribute of the identified code package satisfying the criterion associated with non-trustworthiness, quarantining the identified code package in a sandbox or blocking access to the identified code package.

A computer program product comprising a computer-readable storage medium having instructions recorded thereon for enabling a processor-based system to perform operations, the operations comprising:

• providing an artificial intelligence (AI) prompt, which requests identification of a code package that is written in a programming language and that comprises a designated functionality that resolves a coding problem, to an AI model, which causes the AI model to recommend an identified code package to resolve the coding problem,
• wherein the AI prompt specifies the coding problem, the programming language, and the designated functionality that resolves the coding problem;
• receiving a response to the AI prompt from the AI model, the response comprising a package identifier that identifies the identified code package;
• in response to receipt of the response, which comprises the package identifier that identifies the identified code package, from the AI model, confirming that the identified code package exists;
• based at least on confirmation that the identified code package exists, determining whether a value of an attribute of the identified code package satisfies a criterion associated with non-trustworthiness;
• and
• based at least on the value of the attribute of the identified code package satisfying the criterion associated with non-trustworthiness, quarantining the identified code package in a sandbox or blocking access to the identified code package.