Zero knowledge means no knowledge, and it is under this premise - not leaving any detailed knowledge to the communication partner - that the zero knowledge protocol works.
In the zero knowledge protocol, one of two communicating parties, the prover, has knowledge of a secret, and convinces the communication partner, the verifier, with partial information that he knows the secret. The method is used in encryption and can condition multiple interactive communications between the communicating parties. This is due to the fact that the verifier can request new information
several times, with which the secret is proven, but not revealed. Zero-knowledge is a proof protocol in which the prover only leaves the verifier with enough information to convince the verifier that he knows the secret.