Skip to content
JevDirectory.orgTraffic count unavailable. View site traffic

Jev directory

CommunityJev Tools & CLIs

Laya Website Security Copilot

A local-first Chrome extension that inventories a page's browser-visible attack surface, runs deterministic checks for mixed content, insecure forms and missing headers, and can add typed Laya or official Jev decisions through an optional gateway. Findings carry severity, confidence and evidence.
See highlightsOpen github.com
Stars
0
Language
JavaScript
License
APACHE-2.0
Reading time
2 min
Published by
Community
Author
pateti-hub
Last verified
Oct 2026

pateti-hub/laya-browser-guard

User repository on GitHub

View on GitHub

Privacy-first browser extension for detecting phishing, subscription traps, deceptive interfaces, and dark patterns using multi-signal local AI analysis.

Forks
0
Last pushed
Updated Sep 2026

Repo stats from the GitHub API, cached Oct 2026.

Tagscommunitycatalog-importlogicrwchrome-extensionsecurity
Works forJev for ClassificationJev for ScoringJev for Decision gating
Used inJev for Security & Privacy
Languages
JavaScript
Platforms
Chrome extensionBrowser
Capabilities
Open sourceRuns locally

Highlights

  • Deterministic checks cover scripts, resources, iframes, links, forms, mixed content, credential transport and CSP/HSTS/nosniff/Referrer-Policy headers.
  • The optional gateway sends one structured state and five typed questions to api.typesafe.ai/v1/systemone: two Nouls, two Choices and a Score.
  • Local mode runs the 528 MB nvkudva/laya-web-q8 ONNX model after an explicit user download, checked against published SHA-256 hashes and cached locally.
  • The TypeSafe key stays server-side in the Railway deployment; the extension stores only a deployment-specific gateway token in Chrome.
  • Research Mode allows bounded same-origin JavaScript inspection, and detected secret patterns are automatically redacted before display or export.

Quickstart

bash
npm ci
npm run build

Watch out

Grounded in the truncated repository README and GitHub metadata reviewed on 2026-10-15; the mode list is cut off after Remote, so the three-mode behavior is only partly documented, and no accuracy or latency numbers are published. Apache-2.0, built from source, with the 528 MB model optional.

Questions about this entry

Short answers drawn from the linked project or page.

Reactions & coverage

Posts, threads, and videos about this entry from around the web.

YouTube: Kubernetes admission control with a decision model, fail-closed.

Hacker News: SafeDep Gryph plus Jev checks for compromised coding-agent behavior.

2 pointsDetecting Compromised AI Coding Agents with Jev and GryphRead the thread on Hacker News

X: Jev and Laya on security queries

X: Testing Jev for prompt injection

From the Jev guides

Original write-ups on what this entry is for.

From the Jev AI model community

Posts from builders shipping with Jev right now.

Follow @typesafeai