jev-browser
Browser automation CLI with an interactive terminal and structured commands for agents, using Jev decisions and agent-browser execution.
Overview
jev-browser by Mrlyk is a command-line tool, installed as jevb, for controlling a browser with natural language in an interactive terminal or through structured CLI commands from an AI agent. It uses Jev's fast decisions with agent-browser for execution and ships a companion agent skill installable with npx skills add.
On first run it prompts for a TypeSafe or OpenRouter key (TypeSafe takes priority if both are set), with environment variables usable in CI. By default it connects to your existing Chrome via remote debugging, or can launch headed or headless browsers, reuse sessions, connect to a specific CDP port or auto-connect. The prebuilt npm package supports macOS on Apple Silicon, with source builds for other platforms; documentation is in English and Chinese. It is Apache-2.0 licensed.
What the decision model decides
| Type | Question | Answers |
|---|---|---|
| Choice | Which browser action best advances the natural-language instruction?(inferred) | — |
How it works
- Install jev-browser-cli and log in with a key
- Start the interactive terminal or issue CLI commands
- Jev decides actions; agent-browser executes
- Agents use the companion skill for browser tasks
Strengths
- Works with your existing Chrome session
- Interactive and scriptable modes
- Agent skill included
Caveats
- Prebuilt binary only for macOS Apple Silicon
- Connecting to your real browser exposes logged-in sessions
Getting started
`npm install -g jev-browser-cli`, then run `jevb` and sign in with a TypeSafe or OpenRouter key.At a glance
- Type
- CLI
- Status
- beta
- Language
- TypeScript
- Runtime
- Node CLI/TUI with agent-browser
- Has code
- yes
- License
- Apache-2.0
Use cases and integrations
How curated lists describe it
Browser automation CLI for AI agents, powered by the Jev model's millisecond decisions and near-zero inference costs
logicrw/awesome-jev-projects
Frequently asked questions
What does Jev decide?
The browser actions needed to carry out your instruction.
Which platforms are supported?
Prebuilt for macOS Apple Silicon; others need a source build.
Can agents use it?
Yes, through structured CLI commands and a companion skill.
Related projects
MCP 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 Claude Desktop MCP extension that drives Brave or Chrome with Jev choosing on-page actions and a guard freezing irreversible clicks.
MCP serverbetaTypeScriptChrome MCP server with 50 tools that drives your everyday browser without losing logins, where Jev or a local decider picks each step.
MCP serverbetaStealth browser agent where Jev decides each step in about 0.3 s and CloakBrowser carries it out; MCP server, CLI and Python API.
MCP serverbetaPythonCodexQA browser skill that indexes in-page controls and has Jev pick the operation and target each step, with no screenshots needed.
Agent skillbeta