agentleFS
Sign inStart free
jalbarrangjalbarrang/skills/browser/SKILL.md

browser

Skill2 starsChanged 3 months agoMITView on GitHub

Search the web, read a page as markdown, screenshot pages, interact with a page (click/type/scroll/select/hover/wait), check console errors, and attach via CDP. Use when the user asks to search online, open/visit a URL, scrape readable content, take a screenshot, drive a browser UI, inspect console output, or connect to a running Chrome with remote debugging.

  • Reads credentials

68 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use V3 Core Implementation, V3 DDD Architecture and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • ruvnetSkill

    V3 Core Implementation

    Core module implementation for claude-flow v3. Implements DDD domains, clean architecture patterns, dependency injection, and modular TypeScript codebase with comprehensive testing.

    Also in 15 projects that use this

  • ruvnetSkill

    V3 DDD Architecture

    Domain-Driven Design architecture for claude-flow v3. Implements modular, bounded context architecture with clean separation of concerns and microkernel pattern.

    Also in 15 projects that use this

  • +10 more from ruvnet

The file

Create a free account to mount
Files
SKILL.md

browser

Two thin Node scripts for search + static page fetch, plus direct agent-browser for everything interactive. Screenshots are image files — read them with your own vision (via your file-reading tool); there is no separate analyze step.

Setup

  1. Confirm CLI: which agent-browser (expected /opt/homebrew/bin/agent-browser or similar).
  2. Install browser binaries once: agent-browser install
  3. Optional keys in .agents/browser.json or ~/.agents/browser.json (strings starting with $ pull from env):
{
  "WEB_SEARCH_PROVIDER": "ddg",
  "GOOGLE_CSE_API_KEY": "$GOOGLE_CSE_API_KEY",
  "GOOGLE_CSE_ID": "$GOOGLE_CSE_ID",
  "BRAVE_SEARCH_API_KEY": "$BRAVE_SEARCH_API_KEY"
}

Config key names match the pi extension: GOOGLE_CSE_API_KEY, GOOGLE_CSE_ID, BRAVE_SEARCH_API_KEY, WEB_SEARCH_PROVIDER. DuckDuckGo needs no key and is the default.

Routing

NeedUse
Searchnode <skill>/scripts/search.mjs "<query>"
Static / mostly-HTML page → markdownnode <skill>/scripts/visit.mjs <url>
JS-heavy SPA / login / interactionagent-browser directly (see reference/agent-browser.md)
Screenshotagent-browser open <url> && agent-browser screenshot /tmp/page.png then read the file with vision

Scripts

node <skill>/scripts/search.mjs "<query>" [--engine ddg|google|brave] [--allow a.com,b.com] [--block x.com] [--limit N]
node <skill>/scripts/visit.mjs <url> [--render]

--render shells out to agent-browser open + agent-browser read (rendered DOM). Prefer direct agent-browser for multi-step interaction.

Extraction fidelity

visit.mjs (default) uses a dependency-free extractor: strips script/style/nav/header/footer/aside, then converts headings, paragraphs, links, lists, code, and tables to markdown. It is not Mozilla Readability + Turndown — chrome may leak, nested markup can flatten, and complex layouts lose structure. For JS-rendered content use --render or agent-browser.

References

FileWhen
reference/agent-browser.mdopen/navigate, screenshot, interact, console, CDP

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free