https://aegisgatesecurity.io/lens/Recent content in AegisGate Lens — Free Privacy-First Browser Protection for AI Tools on AegisGate — Secure Every AI InteractionHugo -- gohugo.ioen-ushttps://aegisgatesecurity.io/lens/privacy/Mon, 01 Jan 0001 00:00:00 +0000https://aegisgatesecurity.io/lens/privacy/🛡️ AegisGate Lens v0.5.3 — canonical facts (source: aegisgate-lens repo) 10 AI providers: ChatGPT, Claude, Gemini, Copilot, DuckDuckGo, Perplexity, Mistral, Grok, DeepSeek, Meta AI 5 detection facets: PII (55), secrets (41), XSS (12), compliance (43), ML adversarial (1 model) 504 automated tests: 492/492 Node + 12/12 ML perf/stress 100% adversarial detection (10/10 prompt injection patterns caught by ML model) 100% on-device, zero network egress by default, ML inference included 12 privacy non-negotiables, Apache 2.https://aegisgatesecurity.io/lens/architecture/Mon, 01 Jan 0001 00:00:00 +0000https://aegisgatesecurity.io/lens/architecture/🆕 AegisGate Lens v0.5.3 is LIVE — CWS approved, AMO approved. This page documents the v0.5.3 architecture. 🛡️ AegisGate Lens v0.5.3 — canonical facts (source: aegisgate-lens repo) 10 AI providers: ChatGPT, Claude, Gemini, Copilot, DuckDuckGo, Perplexity, Mistral, Grok, DeepSeek, Meta AI 151 regex patterns across 5 regex facets: PII (65), secrets (41), XSS (12), compliance (24) + 1 ML facet (Char CNN-BiLSTM adversarial detection) 504 automated tests: 492/492 Node + 12/12 ML perf/stress (5/5 stable runs) 2.https://aegisgatesecurity.io/lens/security/Mon, 01 Jan 0001 00:00:00 +0000https://aegisgatesecurity.io/lens/security/🛡️ AegisGate Lens v0.5.3 — canonical facts (source: aegisgate-lens repo) 10 AI providers: ChatGPT, Claude, Gemini, Copilot, DuckDuckGo, Perplexity, Mistral, Grok, DeepSeek, Meta AI 5 detection facets: PII (55), secrets (41), XSS (12), compliance (43), ML adversarial (1 model) 504 automated tests: 492/492 Node + 12/12 ML perf/stress 100% adversarial detection (10/10 prompt injection patterns caught by ML model) 100% on-device, zero network egress by default, ML inference included 12 privacy non-negotiables, Apache 2.https://aegisgatesecurity.io/lens/changelog/Mon, 01 Jan 0001 00:00:00 +0000https://aegisgatesecurity.io/lens/changelog/🛡️ AegisGate Lens v0.5.3 — canonical facts (source: aegisgate-lens repo) 10 AI providers: ChatGPT, Claude, Gemini, Copilot, DuckDuckGo, Perplexity, Mistral, Grok, DeepSeek, Meta AI 5 detection facets: PII (55), secrets (41), XSS (12), compliance (62), IOC + Rampart (54), ML adversarial (1 model) 504 automated tests: 492/492 Node + 12/12 ML perf/stress 100% adversarial detection (10/10 prompt injection patterns caught by ML model) 100% on-device, zero network egress by default, ML inference included 12 privacy non-negotiables, Apache 2.https://aegisgatesecurity.io/lens/terms/Mon, 01 Jan 0001 00:00:00 +0000https://aegisgatesecurity.io/lens/terms/AegisGate Lens — Terms of Service Last updated: 2026-07-29 Effective for: AegisGate Lens v0.5.3 and later Licensor: AegisGate Security, LLC (“AegisGate”, “we”, “us”) Licensee: End user of the AegisGate Lens browser extension (“you”, “your”) These Terms of Service (“ToS”) govern your use of the AegisGate Lens Chrome browser extension. By installing or using the Lens, you agree to these ToS. AegisGate Lens is open-source software licensed under the Apache License 2.https://aegisgatesecurity.io/lens/compare/Mon, 01 Jan 0001 00:00:00 +0000https://aegisgatesecurity.io/lens/compare/🛡️ AegisGate Lens v0.5.3 — canonical facts (source: aegisgate-lens repo) 10 AI providers: ChatGPT, Claude, Gemini, Copilot, DuckDuckGo, Perplexity, Mistral, Grok, DeepSeek, Meta AI 6 detection facets: PII (65), secrets (41), XSS (12), compliance (24), ML adversarial (1 Char CNN-BiLSTM model) — 151 regex patterns + on-device ML 504 automated tests: 492/492 Node + 12/12 ML perf/stress 100% adversarial detection (10/10 prompt injections caught by ML model); 2.31% regex FPR on 6,500 WildChat prompts ~0.