Originally from Vercel Labs, this is a CLI tool for browser automation that talks directly to Chrome via CDP. You install it globally, run `agent-browser install` to grab Chrome, then follow a tight loop: open a URL, snapshot to get element refs like @e1 and @e2, interact with those refs, re-snapshot after changes. It's built for agents that need to drive a real browser without the usual Puppeteer ceremony. The workflow is opinionated but clear, which matters when you're trying to automate forms, scrape dynamic sites, or test flows that need an actual render engine. Feels purpose-built for LLM-driven automation where you want structured element references instead of brittle selectors.

skills.sh installs
542
GitHub stars
26K
last updated
Aug 21, 2026

Install

$ npx -y skills add everyinc/compound-engineering-plugin --skill agent-browser --agent claude-code

Installs into .claude/skills of the current project.

Files

Originally from Vercel Labs, this is a CLI tool for browser automation that talks directly to Chrome via CDP. You install it globally, run `agent-browser install` to grab Chrome, then follow a tight loop: open a URL, snapshot to get element refs like @e1 and @e2, interact with those refs, re-snapshot after changes. It's built for agents that need to drive a real browser without the usual Puppeteer ceremony. The workflow is opinionated but clear, which matters when you're trying to automate forms, scrape dynamic sites, or test flows that need an actual render engine. Feels purpose-built for LLM-driven automation where you want structured element references instead of brittle selectors.

Browse the repository on GitHub

Sponsors

Details

Repository
github.com/everyinc/compound-engineering-plugin
Categories
Testing & QA, AI & Agent Building, Automation & Workflows, Design & UI/UX, CLI & Terminal
First seen
May 16, 2026

More from everyinc/compound-engineering-plugin

All 78
1evoscientist avatarpaper-reviewevoscientist/evoskillsGuides self-review of YOUR OWN academic paper before submission with adversarial stress-testing. Core method: 5-aspect checklist (contribution sufficiency, writ521skills.sh installs
2tldraw avatarwrite-unit-teststldraw/tldrawWriting unit and integration tests for the tldraw SDK. Use when creating new tests, adding test coverage, or fixing failing tests in packages/editor or packages503skills.sh installs
3longbridge avatarlongbridge-risk-analysislongbridge/skillsRisk measurement and stress testing via Longbridge — computes VaR (historical simulation / parametric), CVaR (expected shortfall), max drawdown, Sharpe ratio, C496skills.sh installs
4parcadei avatarcomplete-skillparcadei/continuous-claude-v3A complete skill for E2E testing487skills.sh installs
5pedronauck avatarreactpedronauck/skillsReact 19 development under the React Compiler. Use when writing React components or hooks, deciding whether a useMemo/useCallback/memo belongs in the code, diag473skills.sh installs
6jezweb avataragent-browserjezweb/claude-skillsBrowser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling forms, clicking buttons, taking scr469skills.sh installs