McAfee Endpoint (ePO) Security offers various endpoint security solutions to managed devices. This article provides best practices recommendations to ensure smooth interoperability of Netskope Client and McAfee Endpoint Security installed in a managed device.
We recommend that you read these articles to gain a better understanding of how Client works and its interoperability with 3rd party apps.
This best practices and configurations are based on the following product versions.
We recommend the following configuration requirement to ensure Netskope Client is able to steer traffic to Netskope cloud and also allow McAfee to process their traffic without any conflicts.
Default policies in McAfee ePO does not introduce restrictions on Netskope Client traffic. However, when creating a new policy ensure that the ports 80 and 443 are enabled and allowed in the McAfee Security Firewall rules.
Note
HTTP/HTTPS traffic (via 80 and 443) is enabled and allowed in default firewall policy






Note
If the ports are not allowed or enabled, click the Edit button open the Edit Rule page to select the Allow option listed under Actions and select Enable rule under Status.
In the Netskope tenant WebUI, add McAfee Agent as a certificate pinned app exception and add a set of McAfee URLs as domain exception to the appropriate steering configuration.
Creating symbolic ontologies manually is tedious. Future research focuses on utilizing neural networks to automatically discover and construct robust symbolic rules from raw data. Conclusion
How does a neural network reliably map continuous sensory input (e.g., pixels) to discrete symbols (e.g., "cat") in open-world scenarios? Current methods assume a fixed set of symbols; few handle dynamic symbol creation. Creating symbolic ontologies manually is tedious
Neuro-Symbolic Artificial Intelligence: The State of the Art Current methods assume a fixed set of symbols;
Neuro-symbolic artificial intelligence | European Data Protection Supervisor translating chaotic perceptual inputs into explicit
To explore deeper technical implementations, specific mathematical proofs, or to tailor this research to a particular sub-field,
Neural networks require smooth, continuous mathematical functions to learn via backpropagation. Symbolic logic is discrete, step-based, and non-differentiable. Finding scalable methods to backpropagate gradients through discrete logical operations remains a primary bottleneck.
A system where a neural network generates symbolic rules from raw data. The network acts as an inductive logic programmer, translating chaotic perceptual inputs into explicit, verifiable symbolic code.
Netskope Client is validated to work smoothly with McAfee ePO. To view the validation tests for Netskope Client, see Netskope Client Interoperability
McAfee functions were validated by executing the following tasks: