Agent Browser CLI
Give an agent a headed or headless browser CLI to open pages, snapshot the DOM, and click through flows without a separate MCP server.
Check the source for its package format and setup requirements before installing.
·
When to use this skill
Use Agent Browser CLI when the agent must look at a live page — a staging URL, a docs site, a local dev server — and act on what it sees.
When not to use it
Use Web App Testing when you already have Playwright tests to extend. Use MCP Builder if the team needs a long-lived, authenticated browser tool server.
How to get started
Check the source for its package format and setup requirements before installing.
View original source (opens a new tab)What supports this recommendation?
44/100 provisional content score. This measures documentation, practical detail and source signals. It is not a measured task success rate or a security certification.
- Documentation
- 43/100
- Practical detail
- 40/100
- Source evidence
- 35/100
- Source signals
- 64/100
Model compatibility evidence
- claudeuntested
No model-specific signal or recorded compatibility test was found.
- chatgptuntested
No model-specific signal or recorded compatibility test was found.
- geminiuntested
No model-specific signal or recorded compatibility test was found.
- copilotuntested
No model-specific signal or recorded compatibility test was found.
- llamauntested
No model-specific signal or recorded compatibility test was found.
- perplexityuntested
No model-specific signal or recorded compatibility test was found.
- mistraluntested
No model-specific signal or recorded compatibility test was found.
- grokuntested
No model-specific signal or recorded compatibility test was found.
Inferred compatibility is based on content signals. Check the app, its available tools and setup requirements before running a skill.
What this skill does
This is the successor shape of the old "agent-browser-cli-browser-automation" slug: a CLI the agent shells out to, not a disappearing URL.
Intended uses
- Confirm a deploy actually rendered
- Extract a table from a page with no API
- Reproduce a UI bug the stack trace cannot show
Implementation notes
- Prefer snapshots over screenshots for assertions the agent can read.
- Never send credentials as prompt text; use env files the CLI already understands.
Try it, then tell us how it went
A copied prompt does not tell us whether the task worked. Report your result after checking the output. We collect your tool and outcome, never your files or prompt.