[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7826SCCPFKY7QKDA3EPFBT3":3,"skill-same:01M7826SCCPFKY7QKDA3EPFBT3":56},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":33,"commitSha":34,"skillDirectory":35,"forksCount":36,"repositoryHtmlUrl":37,"repositoryHomepageUrl":38,"ownerHtmlUrl":39,"ownerBlogUrl":6,"ownerDescription":40,"updatedAt":41,"install":42,"resources":55},"01M7826SCCPFKY7QKDA3EPFBT3","agent-browser-automation",null,"Headless browser automation CLI for AI agents using native Rust binary with Chrome DevTools Protocol","Lord1Egypt\u002Fawesome-skill-forge","Lord1Egypt","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F99746359?v=4","01M7592YDNT0XK9WJ0T2MQVPDN","Mohamed Mounir Ali AL Keshita","01M75GRFNHC8VZNWEB8DKDE433",3,"MIT","https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fcommunity\u002Fskills_sh\u002Fs\u002Fskills-sh_aradotso_trending-skills_agent-browser-automation\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: agent-browser-automation\ndescription: Headless browser automation CLI for AI agents using native Rust binary with Chrome DevTools Protocol\ntriggers:\n  - automate browser with agent-browser\n  - use agent-browser for web scraping\n  - browser automation CLI for AI\n  - take screenshot with agent-browser\n  - click and fill forms with agent-browser\n  - agent-browser snapshot accessibility tree\n  - batch browser commands with agent-browser\n  - install agent-browser for automation\n---\n\n# agent-browser\n\n> Skill by [ara.so](https:\u002F\u002Fara.so) — Daily 2026 Skills collection.\n\n`agent-browser` is a headless browser automation CLI built in Rust, designed for AI agents. It wraps Chrome via the Chrome DevTools Protocol (CDP) and exposes a fast, ergonomic command-line interface for navigation, interaction, accessibility snapshots, screenshots, network interception, and more — with no Node.js or Playwright runtime required.\n\n## Installation\n\n### Recommended (npm global)\n```bash\nnpm install -g agent-browser\nagent-browser install  # Download Chrome for Testing (first time only)\n```\n\n### macOS (Homebrew)\n```bash\nbrew install agent-browser\nagent-browser install\n```\n\n### Rust \u002F Cargo\n```bash\ncargo install agent-browser\nagent-browser install\n```\n\n### Local project dependency\n```bash\nnpm install agent-browser\n# Add to package.json scripts or invoke via npx\n```\n\n### Linux (with system dependencies)\n```bash\nagent-browser install --with-deps\n```\n\n## Quick Start\n\n```bash\nagent-browser open https:\u002F\u002Fexample.com\nagent-browser snapshot                        # Accessibility tree with @refs (best for AI)\nagent-browser click @e2                       # Click by ref from snapshot\nagent-browser fill @e3 \"hello@example.com\"   # Fill by ref\nagent-browser get text @e1                    # Get text content\nagent-browser screenshot page.png\nagent-browser close\n```\n\n## Core Commands\n\n### Navigation\n```bash\nagent-browser open \u003Curl>           # Navigate (aliases: goto, navigate)\nagent-browser get url              # Get current URL\nagent-browser get title            # Get page title\nagent-browser close                # Close browser (aliases: quit, exit)\n```\n\n### Accessibility Snapshot (recommended for AI agents)\n```bash\nagent-browser snapshot             # Returns accessibility tree with @ref IDs\nagent-browser snapshot -i          # Interactive \u002F compact mode\n```\n\nSnapshot output includes `@eN` refs you can use directly:\n```\n@e1 [button] \"Submit\"\n@e2 [textbox] \"Email\" value=\"\"\n@e3 [link] \"Sign in\"\n```\n\nThen act on them:\n```bash\nagent-browser fill @e2 \"user@example.com\"\nagent-browser click @e1\n```\n\n### Interaction\n```bash\nagent-browser click \u003Csel>                     # Click element\nagent-browser dblclick \u003Csel>                  # Double-click\nagent-browser fill \u003Csel> \u003Ctext>               # Clear and fill input\nagent-browser type \u003Csel> \u003Ctext>               # Type into element\nagent-browser press \u003Ckey>                     # Press key (Enter, Tab, Control+a)\nagent-browser keyboard type \u003Ctext>            # Type at current focus (real keystrokes)\nagent-browser keyboard inserttext \u003Ctext>      # Insert text without key events\nagent-browser hover \u003Csel>                     # Hover element\nagent-browser select \u003Csel> \u003Cvalue>            # Select dropdown option\nagent-browser check \u003Csel>                     # Check checkbox\nagent-browser uncheck \u003Csel>                   # Uncheck checkbox\nagent-browser scroll down 500                 # Scroll (up\u002Fdown\u002Fleft\u002Fright, optional px)\nagent-browser scroll down --selector \"#feed\"  # Scroll within element\nagent-browser scrollintoview \u003Csel>            # Scroll element into view\nagent-browser drag \u003Csrc> \u003Ctarget>             # Drag and drop\nagent-browser upload \u003Csel> \u002Fpath\u002Ffile.pdf     # Upload file\n```\n\n### Screenshots & PDF\n```bash\nagent-browser screenshot                          # Save to temp dir, print path\nagent-browser screenshot page.png                 # Save to path\nagent-browser screenshot --full page.png          # Full-page screenshot\nagent-browser screenshot --annotate               # Numbered element labels overlay\nagent-browser screenshot --screenshot-dir .\u002Fshots # Custom output directory\nagent-browser screenshot --screenshot-format jpeg --screenshot-quality 80\nagent-browser pdf output.pdf                      # Save page as PDF\n```\n\n### Getting Element Info\n```bash\nagent-browser get text \u003Csel>           # Text content\nagent-browser get html \u003Csel>           # innerHTML\nagent-browser get value \u003Csel>          # Input value\nagent-browser get attr \u003Csel> \u003Cattr>    # Attribute value\nagent-browser get count \u003Csel>          # Count matching elements\nagent-browser get box \u003Csel>            # Bounding box\nagent-browser get styles \u003Csel>         # Computed styles\nagent-browser get cdp-url              # CDP WebSocket URL\n```\n\n### State Checks\n```bash\nagent-browser is visible \u003Csel>\nagent-browser is enabled \u003Csel>\nagent-browser is checked \u003Csel>\n```\n\n### Semantic Locators (find)\n```bash\nagent-browser find role button click --name \"Submit\"\nagent-browser find text \"Sign In\" click\nagent-browser find label \"Email\" fill \"test@example.com\"\nagent-browser find placeholder \"Search...\" fill \"rust\"\nagent-browser find testid \"login-btn\" click\nagent-browser find first \".item\" click\nagent-browser find nth 2 \"a\" text\nagent-browser find role textbox fill \"hello\" --name \"Username\"\n```\n\n**Actions:** `click`, `fill`, `type`, `hover`, `focus`, `check`, `uncheck`, `text`\n\n### Waiting\n```bash\nagent-browser wait \"#modal\"                          # Wait for element visible\nagent-browser wait 2000                              # Wait N milliseconds\nagent-browser wait --text \"Welcome back\"             # Wait for text\nagent-browser wait --url \"**\u002Fdashboard\"              # Wait for URL pattern\nagent-browser wait --load networkidle                # Wait for load state\nagent-browser wait --fn \"window.appReady === true\"   # Wait for JS condition\nagent-browser wait \"#spinner\" --state hidden         # Wait for element to disappear\n```\n\n**Load states:** `load`, `domcontentloaded`, `networkidle`\n\n### JavaScript Eval\n```bash\nagent-browser eval \"document.title\"\nagent-browser eval \"JSON.stringify(window.__STATE__)\"\nagent-browser eval -b \"BASE64_ENCODED_JS\"\necho \"return document.body.innerHTML\" | agent-browser eval --stdin\n```\n\n### Batch Execution (efficient multi-step)\n```bash\necho '[\n  [\"open\", \"https:\u002F\u002Fexample.com\"],\n  [\"snapshot\", \"-i\"],\n  [\"fill\", \"@e2\", \"user@example.com\"],\n  [\"click\", \"@e1\"],\n  [\"screenshot\", \"result.png\"]\n]' | agent-browser batch --json\n\n# Stop on first failure\nagent-browser batch --bail \u003C commands.json\n```\n\n### Tabs & Frames\n```bash\nagent-browser tab                    # List tabs\nagent-browser tab new https:\u002F\u002F...    # New tab with URL\nagent-browser tab 2                  # Switch to tab 2\nagent-browser tab close              # Close current tab\nagent-browser frame \"#my-iframe\"     # Switch into iframe\nagent-browser frame main             # Return to main frame\n```\n\n### Cookies & Storage\n```bash\nagent-browser cookies\nagent-browser cookies set session_id \"abc123\"\nagent-browser cookies clear\n\nagent-browser storage local\nagent-browser storage local set theme dark\nagent-browser storage local clear\nagent-browser storage session set cart '{\"items\":[]}'\n```\n\n### Network\n```bash\nagent-browser network route \"**\u002Fapi\u002Fusers\" --body '{\"users\":[]}'  # Mock response\nagent-browser network route \"**\u002Fads\u002F**\" --abort                    # Block requests\nagent-browser network unroute                                       # Remove all routes\nagent-browser network requests --filter api                        # View requests\nagent-browser network har start\nagent-browser network har stop recording.har\n```\n\n### Browser Settings\n```bash\nagent-browser set viewport 1280 800\nagent-browser set viewport 375 812 2        # With device pixel ratio (retina)\nagent-browser set device \"iPhone 14\"\nagent-browser set geo 37.7749 -122.4194\nagent-browser set offline on\nagent-browser set headers '{\"X-Custom\":\"value\"}'\nagent-browser set credentials admin secret\nagent-browser set media dark\n```\n\n### Auth State\n```bash\nagent-browser state save .\u002Fauth.json    # Save cookies + localStorage\nagent-browser state load .\u002Fauth.json    # Restore auth state\nagent-browser state list                # List saved states\nagent-browser state show auth.json      # Summary of saved state\n```\n\n### Dialogs\n```bash\nagent-browser dialog accept             # Accept alert\u002Fconfirm\u002Fprompt\nagent-browser dialog accept \"My input\"  # Accept prompt with text\nagent-browser dialog dismiss\n```\n\n### Clipboard\n```bash\nagent-browser clipboard read\nagent-browser clipboard write \"Hello, World!\"\nagent-browser clipboard copy           # Ctrl+C current selection\nagent-browser clipboard paste          # Ctrl+V\n```\n\n### Diff & Visual Testing\n```bash\nagent-browser diff snapshot                                  # vs last snapshot\nagent-browser diff snapshot --baseline before.txt            # vs saved file\nagent-browser diff snapshot --selector \"#main\" --compact\nagent-browser diff screenshot --baseline before.png\nagent-browser diff screenshot --baseline b.png -o diff.png\nagent-browser diff url https:\u002F\u002Fv1.example.com https:\u002F\u002Fv2.example.com\nagent-browser diff url https:\u002F\u002Fv1.example.com https:\u002F\u002Fv2.example.com --screenshot\nagent-browser diff url https:\u002F\u002Fv1.example.com https:\u002F\u002Fv2.example.com --selector \"#content\"\n```\n\n### Debug & Profiling\n```bash\nagent-browser trace start trace.zip\nagent-browser trace stop\nagent-browser profiler start\nagent-browser profiler stop profile.json\nagent-browser console                  # View console messages\nagent-browser errors                   # View uncaught JS exceptions\nagent-browser highlight \"#button\"      # Visually highlight element\nagent-browser inspect                  # Open Chrome DevTools\nagent-browser connect 9222             # Connect to existing browser via CDP port\n```\n\n## Common Patterns\n\n### Login flow and save session\n```bash\n#!\u002Fbin\u002Fbash\nagent-browser open https:\u002F\u002Fapp.example.com\u002Flogin\nagent-browser fill \"#email\" \"$LOGIN_EMAIL\"\nagent-browser fill \"#password\" \"$LOGIN_PASSWORD\"\nagent-browser click \"[type=submit]\"\nagent-browser wait --url \"**\u002Fdashboard\"\nagent-browser state save .\u002Fsession.json\n```\n\n### AI agent loop with snapshot-driven interaction\n```bash\n#!\u002Fbin\u002Fbash\nagent-browser open https:\u002F\u002Fapp.example.com\nagent-browser state load .\u002Fsession.json\n\n# Get snapshot, parse @refs, act\nSNAPSHOT=$(agent-browser snapshot)\necho \"$SNAPSHOT\"\n\n# Agent determines @e5 is the search box\nagent-browser fill @e5 \"quarterly report\"\nagent-browser press Enter\nagent-browser wait --load networkidle\nagent-browser snapshot\nagent-browser screenshot results.png\n```\n\n### Batch commands from a script (JSON)\n```bash\ncat > commands.json \u003C\u003C 'EOF'\n[\n  [\"open\", \"https:\u002F\u002Fnews.ycombinator.com\"],\n  [\"wait\", \"--load\", \"networkidle\"],\n  [\"get\", \"title\"],\n  [\"snapshot\"],\n  [\"screenshot\", \"hn.png\"]\n]\nEOF\n\nagent-browser batch --json \u003C commands.json\n```\n\n### Scrape with mocked network\n```bash\nagent-browser open https:\u002F\u002Fapi-heavy-app.example.com\nagent-browser network route \"**\u002Fapi\u002Fslow-endpoint\" --body '{\"data\":\"mocked\"}'\nagent-browser snapshot\nagent-browser network unroute\n```\n\n### Full-page screenshot with annotations\n```bash\nagent-browser open https:\u002F\u002Fexample.com\nagent-browser wait --load networkidle\nagent-browser screenshot --full --annotate annotated.png\n```\n\n### Connect to already-running Chrome\n```bash\n# Start Chrome with remote debugging\ngoogle-chrome --remote-debugging-port=9222 &\n\nagent-browser connect 9222\nagent-browser open https:\u002F\u002Fexample.com\nagent-browser snapshot\n```\n\n### Emulate mobile device\n```bash\nagent-browser set device \"iPhone 14\"\nagent-browser open https:\u002F\u002Fexample.com\nagent-browser screenshot mobile.png\n```\n\n### HAR recording for network analysis\n```bash\nagent-browser open https:\u002F\u002Fexample.com\nagent-browser network har start\nagent-browser click \"#load-data\"\nagent-browser wait --load networkidle\nagent-browser network har stop session.har\n```\n\n## Selector Reference\n\n| Format | Example | Notes |\n|--------|---------|-------|\n| `@ref` | `@e1`, `@e12` | From `snapshot` output — preferred for AI |\n| CSS | `#id`, `.class`, `[attr=val]` | Standard CSS selectors |\n| Text | `\"Sign In\"` | Exact text match |\n| XPath | `\u002F\u002Fbutton[@type='submit']` | Full XPath |\n\n## Troubleshooting\n\n### Chrome not found\n```bash\nagent-browser install              # Downloads Chrome for Testing\nagent-browser install --with-deps  # Linux: also installs system libs\n```\n\n### Element not found \u002F timing issues\n```bash\nagent-browser wait \"#my-element\"              # Wait for visibility first\nagent-browser wait --load networkidle         # Wait for page to settle\nagent-browser wait --fn \"!!document.querySelector('#app')\"\n```\n\n### Selector issues — use snapshot refs instead\n```bash\n# Instead of fragile CSS:\nagent-browser click \".btn.btn-primary.submit-form\"\n\n# Use snapshot refs:\nagent-browser snapshot  # Find @e7 = [button] \"Submit\"\nagent-browser click @e7\n```\n\n### Debug what's on the page\n```bash\nagent-browser screenshot debug.png        # Visual check\nagent-browser snapshot                    # Accessibility tree\nagent-browser console                     # JS console output\nagent-browser errors                      # Uncaught exceptions\nagent-browser eval \"document.readyState\"\n```\n\n### Auth issues between sessions\n```bash\nagent-browser state save .\u002Fauth.json   # After successful login\nagent-browser state load .\u002Fauth.json   # At start of next session\n```\n\n### Handling alerts\u002Fdialogs\n```bash\n# Set up handler BEFORE the action that triggers dialog\nagent-browser dialog accept\nagent-browser click \"#delete-button\"\n```\n\n### Performance — use batch for multi-step workflows\n```bash\n# Slow: one process per command\nagent-browser open https:\u002F\u002Fexample.com\nagent-browser fill \"#q\" \"search\"\nagent-browser click \"#submit\"\n\n# Fast: single process, multiple commands\necho '[[\"open\",\"https:\u002F\u002Fexample.com\"],[\"fill\",\"#q\",\"search\"],[\"click\",\"#submit\"]]' \\\n  | agent-browser batch --json\n```\n","# agent-browser\n\n> Skill by [ara.so](https:\u002F\u002Fara.so) — Daily 2026 Skills collection.\n\n`agent-browser` is a headless browser automation CLI built in Rust, designed for AI agents. It wraps Chrome via the Chrome DevTools Protocol (CDP) and exposes a fast, ergonomic command-line interface for navigation, interaction, accessibility snapshots, screenshots, network interception, and more — with no Node.js or Playwright runtime required.\n\n## Installation\n\n### Recommended (npm global)\n```bash\nnpm install -g agent-browser\nagent-browser install  # Download Chrome for Testing (first time only)\n```\n\n### macOS (Homebrew)\n```bash\nbrew install agent-browser\nagent-browser install\n```\n\n### Rust \u002F Cargo\n```bash\ncargo install agent-browser\nagent-browser install\n```\n\n### Local project dependency\n```bash\nnpm install agent-browser\n# Add to package.json scripts or invoke via npx\n```\n\n### Linux (with system dependencies)\n```bash\nagent-browser install --with-deps\n```\n\n## Quick Start\n\n```bash\nagent-browser open https:\u002F\u002Fexample.com\nagent-browser snapshot                        # Accessibility tree with @refs (best for AI)\nagent-browser click @e2                       # Click by ref from snapshot\nagent-browser fill @e3 \"hello@example.com\"   # Fill by ref\nagent-browser get text @e1                    # Get text content\nagent-browser screenshot page.png\nagent-browser close\n```\n\n## Core Commands\n\n### Navigation\n```bash\nagent-browser open \u003Curl>           # Navigate (aliases: goto, navigate)\nagent-browser get url              # Get current URL\nagent-browser get title            # Get page title\nagent-browser close                # Close browser (aliases: quit, exit)\n```\n\n### Accessibility Snapshot (recommended for AI agents)\n```bash\nagent-browser snapshot             # Returns accessibility tree with @ref IDs\nagent-browser snapshot -i          # Interactive \u002F compact mode\n```\n\nSnapshot output includes `@eN` refs you can use directly:\n```\n@e1 [button] \"Submit\"\n@e2 [textbox] \"Email\" value=\"\"\n@e3 [link] \"Sign in\"\n```\n\nThen act on them:\n```bash\nagent-browser fill @e2 \"user@example.com\"\nagent-browser click @e1\n```\n\n### Interaction\n```bash\nagent-browser click \u003Csel>                     # Click element\nagent-browser dblclick \u003Csel>                  # Double-click\nagent-browser fill \u003Csel> \u003Ctext>               # Clear and fill input\nagent-browser type \u003Csel> \u003Ctext>               # Type into element\nagent-browser press \u003Ckey>                     # Press key (Enter, Tab, Control+a)\nagent-browser keyboard type \u003Ctext>            # Type at current focus (real keystrokes)\nagent-browser keyboard inserttext \u003Ctext>      # Insert text without key events\nagent-browser hover \u003Csel>                     # Hover element\nagent-browser select \u003Csel> \u003Cvalue>            # Select dropdown option\nagent-browser check \u003Csel>                     # Check checkbox\nagent-browser uncheck \u003Csel>                   # Uncheck checkbox\nagent-browser scroll down 500                 # Scroll (up\u002Fdown\u002Fleft\u002Fright, optional px)\nagent-browser scroll down --selector \"#feed\"  # Scroll within element\nagent-browser scrollintoview \u003Csel>            # Scroll element into view\nagent-browser drag \u003Csrc> \u003Ctarget>             # Drag and drop\nagent-browser upload \u003Csel> \u002Fpath\u002Ffile.pdf     # Upload file\n```\n\n### Screenshots & PDF\n```bash\nagent-browser screenshot                          # Save to temp dir, print path\nagent-browser screenshot page.png                 # Save to path\nagent-browser screenshot --full page.png          # Full-page screenshot\nagent-browser screenshot --annotate               # Numbered element labels overlay\nagent-browser screenshot --screenshot-dir .\u002Fshots # Custom output directory\nagent-browser screenshot --screenshot-format jpeg --screenshot-quality 80\nagent-browser pdf output.pdf                      # Save page as PDF\n```\n\n### Getting Element Info\n```bash\nagent-browser get text \u003Csel>           # Text content\nagent-browser get html \u003Csel>           # innerHTML\nagent-browser get value \u003Csel>          # Input value\nagent-browser get attr \u003Csel> \u003Cattr>    # Attribute value\nagent-browser get count \u003Csel>          # Count matching elements\nagent-browser get box \u003Csel>            # Bounding box\nagent-browser get styles \u003Csel>         # Computed styles\nagent-browser get cdp-url              # CDP WebSocket URL\n```\n\n### State Checks\n```bash\nagent-browser is visible \u003Csel>\nagent-browser is enabled \u003Csel>\nagent-browser is checked \u003Csel>\n```\n\n### Semantic Locators (find)\n```bash\nagent-browser find role button click --name \"Submit\"\nagent-browser find text \"Sign In\" click\nagent-browser find label \"Email\" fill \"test@example.com\"\nagent-browser find placeholder \"Search...\" fill \"rust\"\nagent-browser find testid \"login-btn\" click\nagent-browser find first \".item\" click\nagent-browser find nth 2 \"a\" text\nagent-browser find role textbox fill \"hello\" --name \"Username\"\n```\n\n**Actions:** `click`, `fill`, `type`, `hover`, `focus`, `check`, `uncheck`, `text`\n\n### Waiting\n```bash\nagent-browser wait \"#modal\"                          # Wait for element visible\nagent-browser wait 2000                              # Wait N milliseconds\nagent-browser wait --text \"Welcome back\"             # Wait for text\nagent-browser wait --url \"**\u002Fdashboard\"              # Wait for URL pattern\nagent-browser wait --load networkidle                # Wait for load state\nagent-browser wait --fn \"window.appReady === true\"   # Wait for JS condition\nagent-browser wait \"#spinner\" --state hidden         # Wait for element to disappear\n```\n\n**Load states:** `load`, `domcontentloaded`, `networkidle`\n\n### JavaScript Eval\n```bash\nagent-browser eval \"document.title\"\nagent-browser eval \"JSON.stringify(window.__STATE__)\"\nagent-browser eval -b \"BASE64_ENCODED_JS\"\necho \"return document.body.innerHTML\" | agent-browser eval --stdin\n```\n\n### Batch Execution (efficient multi-step)\n```bash\necho '[\n  [\"open\", \"https:\u002F\u002Fexample.com\"],\n  [\"snapshot\", \"-i\"],\n  [\"fill\", \"@e2\", \"user@example.com\"],\n  [\"click\", \"@e1\"],\n  [\"screenshot\", \"result.png\"]\n]' | agent-browser batch --json\n\n# Stop on first failure\nagent-browser batch --bail \u003C commands.json\n```\n\n### Tabs & Frames\n```bash\nagent-browser tab                    # List tabs\nagent-browser tab new https:\u002F\u002F...    # New tab with URL\nagent-browser tab 2                  # Switch to tab 2\nagent-browser tab close              # Close current tab\nagent-browser frame \"#my-iframe\"     # Switch into iframe\nagent-browser frame main             # Return to main frame\n```\n\n### Cookies & Storage\n```bash\nagent-browser cookies\nagent-browser cookies set session_id \"abc123\"\nagent-browser cookies clear\n\nagent-browser storage local\nagent-browser storage local set theme dark\nagent-browser storage local clear\nagent-browser storage session set cart '{\"items\":[]}'\n```\n\n### Network\n```bash\nagent-browser network route \"**\u002Fapi\u002Fusers\" --body '{\"users\":[]}'  # Mock response\nagent-browser network route \"**\u002Fads\u002F**\" --abort                    # Block requests\nagent-browser network unroute                                       # Remove all routes\nagent-browser network requests --filter api                        # View requests\nagent-browser network har start\nagent-browser network har stop recording.har\n```\n\n### Browser Settings\n```bash\nagent-browser set viewport 1280 800\nagent-browser set viewport 375 812 2        # With device pixel ratio (retina)\nagent-browser set device \"iPhone 14\"\nagent-browser set geo 37.7749 -122.4194\nagent-browser set offline on\nagent-browser set headers '{\"X-Custom\":\"value\"}'\nagent-browser set credentials admin secret\nagent-browser set media dark\n```\n\n### Auth State\n```bash\nagent-browser state save .\u002Fauth.json    # Save cookies + localStorage\nagent-browser state load .\u002Fauth.json    # Restore auth state\nagent-browser state list                # List saved states\nagent-browser state show auth.json      # Summary of saved state\n```\n\n### Dialogs\n```bash\nagent-browser dialog accept             # Accept alert\u002Fconfirm\u002Fprompt\nagent-browser dialog accept \"My input\"  # Accept prompt with text\nagent-browser dialog dismiss\n```\n\n### Clipboard\n```bash\nagent-browser clipboard read\nagent-browser clipboard write \"Hello, World!\"\nagent-browser clipboard copy           # Ctrl+C current selection\nagent-browser clipboard paste          # Ctrl+V\n```\n\n### Diff & Visual Testing\n```bash\nagent-browser diff snapshot                                  # vs last snapshot\nagent-browser diff snapshot --baseline before.txt            # vs saved file\nagent-browser diff snapshot --selector \"#main\" --compact\nagent-browser diff screenshot --baseline before.png\nagent-browser diff screenshot --baseline b.png -o diff.png\nagent-browser diff url https:\u002F\u002Fv1.example.com https:\u002F\u002Fv2.example.com\nagent-browser diff url https:\u002F\u002Fv1.example.com https:\u002F\u002Fv2.example.com --screenshot\nagent-browser diff url https:\u002F\u002Fv1.example.com https:\u002F\u002Fv2.example.com --selector \"#content\"\n```\n\n### Debug & Profiling\n```bash\nagent-browser trace start trace.zip\nagent-browser trace stop\nagent-browser profiler start\nagent-browser profiler stop profile.json\nagent-browser console                  # View console messages\nagent-browser errors                   # View uncaught JS exceptions\nagent-browser highlight \"#button\"      # Visually highlight element\nagent-browser inspect                  # Open Chrome DevTools\nagent-browser connect 9222             # Connect to existing browser via CDP port\n```\n\n## Common Patterns\n\n### Login flow and save session\n```bash\n#!\u002Fbin\u002Fbash\nagent-browser open https:\u002F\u002Fapp.example.com\u002Flogin\nagent-browser fill \"#email\" \"$LOGIN_EMAIL\"\nagent-browser fill \"#password\" \"$LOGIN_PASSWORD\"\nagent-browser click \"[type=submit]\"\nagent-browser wait --url \"**\u002Fdashboard\"\nagent-browser state save .\u002Fsession.json\n```\n\n### AI agent loop with snapshot-driven interaction\n```bash\n#!\u002Fbin\u002Fbash\nagent-browser open https:\u002F\u002Fapp.example.com\nagent-browser state load .\u002Fsession.json\n\n# Get snapshot, parse @refs, act\nSNAPSHOT=$(agent-browser snapshot)\necho \"$SNAPSHOT\"\n\n# Agent determines @e5 is the search box\nagent-browser fill @e5 \"quarterly report\"\nagent-browser press Enter\nagent-browser wait --load networkidle\nagent-browser snapshot\nagent-browser screenshot results.png\n```\n\n### Batch commands from a script (JSON)\n```bash\ncat > commands.json \u003C\u003C 'EOF'\n[\n  [\"open\", \"https:\u002F\u002Fnews.ycombinator.com\"],\n  [\"wait\", \"--load\", \"networkidle\"],\n  [\"get\", \"title\"],\n  [\"snapshot\"],\n  [\"screenshot\", \"hn.png\"]\n]\nEOF\n\nagent-browser batch --json \u003C commands.json\n```\n\n### Scrape with mocked network\n```bash\nagent-browser open https:\u002F\u002Fapi-heavy-app.example.com\nagent-browser network route \"**\u002Fapi\u002Fslow-endpoint\" --body '{\"data\":\"mocked\"}'\nagent-browser snapshot\nagent-browser network unroute\n```\n\n### Full-page screenshot with annotations\n```bash\nagent-browser open https:\u002F\u002Fexample.com\nagent-browser wait --load networkidle\nagent-browser screenshot --full --annotate annotated.png\n```\n\n### Connect to already-running Chrome\n```bash\n# Start Chrome with remote debugging\ngoogle-chrome --remote-debugging-port=9222 &\n\nagent-browser connect 9222\nagent-browser open https:\u002F\u002Fexample.com\nagent-browser snapshot\n```\n\n### Emulate mobile device\n```bash\nagent-browser set device \"iPhone 14\"\nagent-browser open https:\u002F\u002Fexample.com\nagent-browser screenshot mobile.png\n```\n\n### HAR recording for network analysis\n```bash\nagent-browser open https:\u002F\u002Fexample.com\nagent-browser network har start\nagent-browser click \"#load-data\"\nagent-browser wait --load networkidle\nagent-browser network har stop session.har\n```\n\n## Selector Reference\n\n| Format | Example | Notes |\n|--------|---------|-------|\n| `@ref` | `@e1`, `@e12` | From `snapshot` output — preferred for AI |\n| CSS | `#id`, `.class`, `[attr=val]` | Standard CSS selectors |\n| Text | `\"Sign In\"` | Exact text match |\n| XPath | `\u002F\u002Fbutton[@type='submit']` | Full XPath |\n\n## Troubleshooting\n\n### Chrome not found\n```bash\nagent-browser install              # Downloads Chrome for Testing\nagent-browser install --with-deps  # Linux: also installs system libs\n```\n\n### Element not found \u002F timing issues\n```bash\nagent-browser wait \"#my-element\"              # Wait for visibility first\nagent-browser wait --load networkidle         # Wait for page to settle\nagent-browser wait --fn \"!!document.querySelector('#app')\"\n```\n\n### Selector issues — use snapshot refs instead\n```bash\n# Instead of fragile CSS:\nagent-browser click \".btn.btn-primary.submit-form\"\n\n# Use snapshot refs:\nagent-browser snapshot  # Find @e7 = [button] \"Submit\"\nagent-browser click @e7\n```\n\n### Debug what's on the page\n```bash\nagent-browser screenshot debug.png        # Visual check\nagent-browser snapshot                    # Accessibility tree\nagent-browser console                     # JS console output\nagent-browser errors                      # Uncaught exceptions\nagent-browser eval \"document.readyState\"\n```\n\n### Auth issues between sessions\n```bash\nagent-browser state save .\u002Fauth.json   # After successful login\nagent-browser state load .\u002Fauth.json   # At start of next session\n```\n\n### Handling alerts\u002Fdialogs\n```bash\n# Set up handler BEFORE the action that triggers dialog\nagent-browser dialog accept\nagent-browser click \"#delete-button\"\n```\n\n### Performance — use batch for multi-step workflows\n```bash\n# Slow: one process per command\nagent-browser open https:\u002F\u002Fexample.com\nagent-browser fill \"#q\" \"search\"\nagent-browser click \"#submit\"\n\n# Fast: single process, multiple commands\necho '[[\"open\",\"https:\u002F\u002Fexample.com\"],[\"fill\",\"#q\",\"search\"],[\"click\",\"#submit\"]]' \\\n  | agent-browser batch --json\n```\n",{"name":5,"triggers":24,"description":7},[25,26,27,28,29,30,31,32],"automate browser with agent-browser","use agent-browser for web scraping","browser automation CLI for AI","take screenshot with agent-browser","click and fill forms with agent-browser","agent-browser snapshot accessibility tree","batch browser commands with agent-browser","install agent-browser for automation","master","4aa004da461d820f4ea5499e58eb8d196cf99016","community\u002Fskills_sh\u002Fs\u002Fskills-sh_aradotso_trending-skills_agent-browser-automation",0,"https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge","https:\u002F\u002Fskill-explorer-rho.vercel.app","https:\u002F\u002Fgithub.com\u002FLord1Egypt","Unbreakable. Unstoppable. Unmatched⚡.\r\n Speaker of machine tongues 🤖  | World-builder of decentralized dreams 🌐 | Defender of the digital realm🛡️.","2026-06-10 08:21:01",{"source":8,"skillName":5,"commands":43,"prompt":53,"zipAvailable":54},[44,47,50],{"tool":45,"command":46},"npx","npx skills add Lord1Egypt\u002Fawesome-skill-forge --skill agent-browser-automation",{"tool":48,"command":49},"bunx","bunx skills add Lord1Egypt\u002Fawesome-skill-forge --skill agent-browser-automation",{"tool":51,"command":52},"pnpm","pnpm dlx skills add Lord1Egypt\u002Fawesome-skill-forge --skill agent-browser-automation","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fcommunity\u002Fskills_sh\u002Fs\u002Fskills-sh_aradotso_trending-skills_agent-browser-automation\u002FSKILL.md",true,[],[57,64,71,78,85,92],{"skillNo":58,"name":59,"nameZh":6,"description":60,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":61,"categories":62,"keywords":63,"publishedAt":19},"01M785WKJB99G3STR1NF9NRTJW","youtube-content","YouTube transcripts to summaries, threads, blogs.","https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fskills\u002Fyoutube-content\u002FSKILL.md",[],[],{"skillNo":65,"name":66,"nameZh":6,"description":67,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":68,"categories":69,"keywords":70,"publishedAt":19},"01M785WEFHB0G2S9YA6EPR3H0E","xurl","X\u002FTwitter via xurl CLI: post, search, DM, media, v2 API.","https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fskills\u002Fxurl\u002FSKILL.md",[],[],{"skillNo":72,"name":73,"nameZh":6,"description":74,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":75,"categories":76,"keywords":77,"publishedAt":19},"01M785W98EDN4EFJXJPMQEHCRC","touchdesigner-mcp","Control a running TouchDesigner instance via twozero MCP — create operators, set parameters, wire connections, execute Python, build real-time visuals. 36 native tools.","https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fskills\u002Ftouchdesigner-mcp\u002FSKILL.md",[],[],{"skillNo":79,"name":80,"nameZh":6,"description":81,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":82,"categories":83,"keywords":84,"publishedAt":19},"01M785W5S9FP3QBH2V191E025P","test-driven-development","TDD: enforce RED-GREEN-REFACTOR, tests before code.","https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fskills\u002Ftest-driven-development\u002FSKILL.md",[],[],{"skillNo":86,"name":87,"nameZh":6,"description":88,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":89,"categories":90,"keywords":91,"publishedAt":19},"01M785W2AS08NSSH9YATCV6FJB","teams-meeting-pipeline","Operate the Teams meeting summary pipeline via Hermes CLI — summarize meetings, inspect pipeline status, replay jobs, manage Microsoft Graph subscriptions.","https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fskills\u002Fteams-meeting-pipeline\u002FSKILL.md",[],[],{"skillNo":93,"name":94,"nameZh":6,"description":95,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":96,"categories":97,"keywords":98,"publishedAt":19},"01M785VXKW4T34BTP1SCHW1R7F","systematic-debugging","4-phase root cause debugging: understand bugs before fixing.","https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fskills\u002Fsystematic-debugging\u002FSKILL.md",[],[]]