AccuKnox Prompt Firewall

Scans ChatGPT and Claude prompts and responses through AccuKnox Prompt Firewall for security and compliance

AccuKnox Prompt Firewall protects your organization when employees use AI assistants like ChatGPT and Claude. It inspects prompts before they are sent and responses before they are shown, applying your AccuKnox Prompt Firewall security and compliance policies in real time. KEY FEATURES • Real-time prompt scanning — every message you send to ChatGPT or Claude is checked against your organization's policies before it leaves the browser. • Response scanning — AI responses are inspected before they are displayed, so sensitive or non-compliant content can be blocked or flagged. • Policy-driven verdicts — content is allowed, blocked, or flagged with a warning based on the policy returned by your AccuKnox firewall. • Activity logs — review a local history of scanned prompts and responses and their verdicts. • Centrally manageable — settings can be deployed and enforced by administrators through managed storage (MDM/Intune), making it suitable for enterprise rollout. HOW IT WORKS 1. Install the extension and enter the API key provided by your AccuKnox tenant. 2. When you use ChatGPT or Claude, the extension sends prompt and response content to your AccuKnox Prompt Firewall endpoint for evaluation. 3. Based on the policy verdict, content is allowed through, blocked, or shown with a warning. REQUIREMENTS An active AccuKnox account and API key are required. This extension is intended for organizations that use the AccuKnox security platform. PRIVACY Prompt and response content is sent only to the AccuKnox firewall endpoint associated with your account for the sole purpose of security and compliance scanning. See our privacy policy for details.