agent-browser
agent-browser lets you script repetitive web interactions—form submissions, element clicks, page navigation—directly through Claude Code. Built for end-to-end encrypted access, it brings browser automation to your workflow without leaving SkillFed's environment.
agent-browser enables you to automate browser testing by scripting repetitive web interactions directly through Claude Code. You can simulate user actions like clicking elements, filling forms, and navigating pages, then verify results by comparing page states visually and textually. The tool integrates seamlessly into SkillFed's environment for end-to-end encrypted access to your automation workflows.
AI-generated summary based on this skill's SKILL.md
Install
slopus/happy/agent-browser · repository language: TypeScript
git clone https://github.com/slopus/happy
cp -r happy/.agents/skills/agent-browser ~/.claude/skills/agent-browserFrequently asked questions
AI-generated answers based on this skill's SKILL.md and metadata
How do I automate browser testing with agent-browser?
agent-browser enables you to automate browser testing by scripting repetitive web interactions directly through Claude Code. You can simulate user actions like clicking elements, filling forms, and navigating pages, then verify results by comparing page states visually and textually. The tool integrates seamlessly into SkillFed's environment for end-to-end encrypted access to your automation workflows.
Can agent-browser click and fill forms automatically?
Yes, agent-browser is designed to click elements and fill forms automatically as part of its core functionality. You can script form submissions, element interactions, and navigation sequences through Claude Code. This automation capability makes agent-browser ideal for handling repetitive browser tasks without manual intervention.
What can agent-browser extract from websites?
agent-browser lets you extract data and capture page state through snapshots and screenshots. You can retrieve text from pages, take screenshots of web pages programmatically, and compare website screenshots to detect changes. This data extraction capability supports web scraping and state verification workflows within your automation scripts.
How does agent-browser handle browser session management?
agent-browser persists authentication state and manages browser sessions across multiple tasks, allowing you to maintain logged-in states and session continuity. This means you can automate workflows that require login persistence, navigate between authenticated pages, and execute multi-step interactions without re-authenticating between actions.
Can agent-browser take screenshots and compare page states?
Yes, agent-browser can take screenshots of web pages programmatically and compare page states both visually and textually. This functionality lets you detect changes on websites, verify that automation actions produced expected results, and capture the state of pages at different points in your workflow.
What license does agent-browser use?
agent-browser is released under the MIT license, which permits free use, modification, and distribution with minimal restrictions. This open-source licensing makes agent-browser accessible for both personal and commercial automation projects within SkillFed's environment.
SKILL.md
rendered from the published skill — quoted content, verbatim
Browser Automation with agent-browser
Core Workflow
Every browser automation follows this pattern:
- Navigate:
agent-browser open <url> - Snapshot:
agent-browser snapshot -i(get element refs like@e1,@e2) - Interact: Use refs to click, fill, select
- Re-snapshot: After navigation or DOM changes, get fresh refs
agent-browser open https://example.com/form
agent-browser snapshot -i
# Output: @e1 [input type="email"], @e2 [input type="password"], @e3 [button] "Submit"
agent-browser fill @e1 "user@example.com"
agent-browser fill @e2 "password123"
agent-browser click @e3
agent-browser wait --load networkidle
agent-browser snapshot -i # Check result
Command Chaining
Commands can be chained with && in a single shell invocation. The browser persists between commands via a background daemon, so chaining is safe and more efficient than separate calls.
```bash
Chain open + wait + snapshot in one call
agent-browser open
(truncated - see the full file via the links below)
Read as markdown · JSON record · Browse the source repository
File tree — 1 file
.agents/skills/agent-browser/SKILL.md