agentleFS
Sign inStart free
Layneformalized225Layneformalized225/ai-cofounder/skills/web-browser/SKILL.md

web-browser

Skill1 starsChanged 8 months agoMITView on GitHub

Browse the web using agent-browser CLI with Lightpanda engine. Read pages, extract content, interact with elements.

8 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

5 rules. 0 are common, 2 are distinctive, and 3 are very specific to this project.

  • 0

    Common

    Most of the 6 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Use agent-browser CLI via exec for all web browsing tasks.

    line 9
  2. Distinctive:

    Default engine: Lightpanda (fast, low-RAM).

    line 10
  3. Very specific:

    If need to interact: exec("agent-browser snapshot -i") - get interactive elements with refs

    line 59
  4. Distinctive:

    Use Chrome engine only when Lightpanda fails on a complex site or you need a screenshot.

    line 75
  5. Very specific:

    If a site crashes with Lightpanda, retry with --engine chrome

    line 83

“Common” is measured: kept by at least 6 in 10 of the 6 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Web Browser (agent-browser + Lightpanda)

Use agent-browser CLI via exec for all web browsing tasks. Default engine: Lightpanda (fast, low-RAM). Falls back to Chrome if needed.

Quick Reference

Open a page

agent-browser open https://example.com

Get page content (text snapshot)

# Full content with text (for reading pages)
agent-browser snapshot

# Interactive elements only (for finding buttons/links/inputs to click)
agent-browser snapshot -i

# Structure only, no text (for understanding page layout)
agent-browser snapshot -c

# Scoped to a CSS selector
agent-browser snapshot -s "#main-content"

Interact with elements

Snapshots return refs like @e1, @e2. Use them to interact:

agent-browser click @e3          # Click a link/button
agent-browser fill @e5 "text"    # Fill an input field
agent-browser type @e5 "text"    # Type without clearing
agent-browser select @e7 "value" # Select dropdown option
agent-browser press @e5 Enter    # Press key
agent-browser hover @e2          # Hover
agent-browser scroll @e1         # Scroll to element

Navigation

agent-browser back               # Go back
agent-browser forward            # Go forward
agent-browser open <new-url>     # Navigate to new URL

Typical Workflow

  1. exec("agent-browser open https://example.com") - open page
  2. exec("agent-browser snapshot") - read page content
  3. If need to interact: exec("agent-browser snapshot -i") - get interactive elements with refs
  4. exec("agent-browser click @e3") - click element
  5. exec("agent-browser snapshot") - check result

Engine Selection

# Lightpanda (default, fast, low RAM, no screenshots)
agent-browser --engine lightpanda open https://example.com

# Chrome (fallback for complex sites or screenshots)
agent-browser --engine chrome open https://example.com
agent-browser screenshot              # Only with Chrome engine
agent-browser screenshot --annotate   # Annotated screenshot

Use Chrome engine only when Lightpanda fails on a complex site or you need a screenshot.

Notes

  • Lightpanda: no screenshots, no visual rendering, but 10x faster and 10x less RAM
  • Refs (@e1) persist across exec calls — daemon stays alive between commands
  • snapshot — full text for reading; snapshot -i — refs for clicking; snapshot -c — layout only
  • Google and some sites block headless browsers from cloud IPs (empty page) — use web_search instead
  • If a site crashes with Lightpanda, retry with --engine chrome

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