jev-chrome-extension
Chrome side-panel extension where Jev drives the current tab toward a typed goal, or answers quick questions about highlighted text.
Overview
Jev for Chrome is a side-panel extension with two modes. In Drive, you type a goal such as search Wikipedia for espresso and open the article, and Jev clicks, types and scrolls the current tab until done. It reads the page as a numbered list of links, buttons and fields (the numbers appear on the page), asks TypeSafe which to use next, performs it and repeats, with each step shown in the panel with Jev's confidence. Step mode pauses after each action, Stop ends a run, and Copy trace exports the run as JSON.
Because Jev only chooses and never writes, any text to type must be given in quotes in the goal, and Jev picks which quoted value fits each field. The second mode lets you highlight text and ask quick yes/no, pick-one or rating questions. It installs as an unpacked extension from a release zip with your own TypeSafe key, tested with a key-check button, and opens with Alt+J.
What the decision model decides
| Type | Question | Answers |
|---|---|---|
| Choice | Which numbered link, button or field should be used next? | — |
| Choice | Which quoted value belongs in this field? | — |
| Yes / no | User's yes/no question about highlighted text | yes · no |
How it works
- Open a site and type a goal with quoted text values
- Number the page's interactive elements
- Jev picks the next element; extension acts
- Repeat with confidence shown; step, stop or copy trace
Strengths
- Works on the tab you are looking at
- Step-by-step mode for watching decisions
- No coding to install
Caveats
- Acts in your logged-in browser
- Typed text must be supplied in quotes
Getting started
Download jev-chrome-extension.zip, load it unpacked in chrome://extensions, then add your TypeSafe key in Options.At a glance
- Type
- Plugin
- Status
- beta
- Language
- JavaScript
- Runtime
- Chrome extension
- Has code
- yes
Use cases and integrations
How curated lists describe it
1. Open any website. 2. Click the Jev icon. The side panel opens on \*\*Drive\*\*. 3. Type a goal, e.g. \*Search Wikipedia for "espresso" and open the article\*, and press \*\*Run\*\*.
logicrw/awesome-jev-projects
Frequently asked questions
What does Jev decide?
Which page element to act on next, and which quoted text belongs in each field.
Can it type arbitrary text?
Only values you put in quotes in the goal.
Can I watch step by step?
Yes, Step mode pauses after each action.
Related projects
RAM-efficient Rust coding-agent harness, listed as using Jev decisions for memory pruning, browser navigation and voice routing.
ServiceproductionRustMCP server and libraries that drive Chrome over CDP and hand a decision model like Jev a bounded menu of page actions.
ProjectproductionTypeScript- Jev Browser★ 335
Headless browser automation via MCP server, CLI or library: Jev picks one action per step and scores goal-met and stuck probabilities.
MCP serverbetaTypeScript Local browser agent that picks each step's operation and element with a Jev-shaped model such as a fine-tuned Laya, no cloud needed.
MCP serverbetaPythonClaude Desktop MCP extension that drives Brave or Chrome with Jev choosing on-page actions and a guard freezing irreversible clicks.
MCP serverbetaTypeScriptBrowser-agent benchmark with five timed events from one prompt, plus a ~200-line DevTools agent where Jev picks each action.
ProjectbetaJavaScript