We are thrilled to announce the launch of Wozway, a cutting-edge solution by DefendAI designed to enhance the security and efficiency of communication between your chat client applications and various LLM (Large Language Model) cloud services.
Wozway is an AI Firewall tht provides a secure and efficient way to manage communication between your chat client applications and LLM cloud services like OpenAI, Groq, Gemini, Anthropic, Perplexity, and more. By running a simple script, you can start a tenant that proxies your LLM requests and responses through a policy-driven LLM proxy, adding an extra layer of security and abstraction.
Secure Proxying: Wozway securely proxies requests to multiple LLM providers, ensuring that your data remains protected.
Policy-Based Enforcement: Utilize API or UI to enforce policies, including regex, anonymization, code, phishing URL, prompt, and response detection.
Similarity Search: Leverage a database of millions of malicious prompts to block or alert administrators.
Monitoring and Dashboard: Access incident, activities, and dashboard features for comprehensive monitoring of usage across all Generative AI apps.
Wozway runs alongside client apps and intelligently re-routes only LLM-bound traffic to DefendAI Cloud for policy enforcement and subsequent forwarding to LLM providers like OpenAI, Anthropic, Groq, etc. All other traffic is routed without any interference to its intended destination. Wozway can intercept both prompts and responses for running full security checks.
Clone the Repository: git clone https://github.com/Defend-AI-Tech-Inc/wozway.git
Install Dependencies: pip install -r requirements.txt
Start the Tenant: Run python start_tenant.py to set up your cloud account and domain.
The Wozway SDK can be installed using pip and provides a simple interface to interact with the service.
For detailed documentation, installation guides, and support, visit our .
We are excited to see how Wozway will enhance the security and efficiency of your AI-driven applications. Join us in this journey to make AI safer and more reliable!
The DEFENDAI Solution – Comprehensive Secure Gen AI
