Code signing is the mobile code security model that relies only on trust. Mobile code is a type of software that can be transferred from one system to another and executed without installation or compilation. Mobile code can be used for various purposes, such as web applications, applets, scripts, macros, etc. Mobile code can also pose various security risks, such as malicious code, unauthorized access, data leakage, etc. Mobile code security models are the techniques that are used to protect the systems and users from the threats of mobile code. Code signing is a mobile code security model that relies only on trust, which means that the security of the mobile code depends on the reputation and credibility of the code provider. Code signing works as follows:

The code provider has a pair of public and private keys, and obtains a digital certificate from a trusted third party, such as a certificate authority (CA), that binds the public key to the identity of the code provider.

The code provider signs the mobile code with its private key and attaches the digital certificate to the mobile code.

The code consumer receives the mobile code and verifies the signature and the certificate with the public key of the code provider and the CA, respectively.

The code consumer decides whether to trust and execute the mobile code based on the identity and reputation of the code provider.

Code signing relies only on trust because it does not enforce any security restrictions or controls on the mobile code, but rather leaves the decision to the code consumer. Code signing also does not guarantee the quality or functionality of the mobile code, but rather the authenticity and integrity of the code provider. Code signing can be effective if the code consumer knows and trusts the code provider, and if the code provider follows the security standards and best practices. However, code signing can also be ineffective if the code consumer is unaware or careless of the code provider, or if the code provider is compromised or malicious.

The other options are not mobile code security models that rely only on trust, but rather on other techniques that limit or isolate the mobile code. Class authentication is a mobile code security model that verifies the permissions and capabilities of the mobile code based on its class or type, and allows or denies the execution of the mobile code accordingly. Sandboxing is a mobile code security model that executes the mobile code in a separate and restricted environment, and prevents the mobile code from accessing or affecting the system resources or data. Type safety is a mobile code security model that checks the validity and consistency of the mobile code, and prevents the mobile code from performing illegal or unsafe operations.