diff --git a/plugins/compound-engineering/.claude-plugin/plugin.json b/plugins/compound-engineering/.claude-plugin/plugin.json
index f4688b2..1cd5183 100644
--- a/plugins/compound-engineering/.claude-plugin/plugin.json
+++ b/plugins/compound-engineering/.claude-plugin/plugin.json
@@ -1,7 +1,7 @@
 {
   "name": "compound-engineering",
-  "version": "2.23.1",
-  "description": "AI-powered development tools. 27 agents, 21 commands, 13 skills, 2 MCP servers for code review, research, design, and workflow automation.",
+  "version": "2.24.0",
+  "description": "AI-powered development tools. 27 agents, 20 commands, 14 skills, 1 MCP server for code review, research, design, and workflow automation.",
   "author": {
     "name": "Kieran Klaassen",
     "email": "kieran@every.to",
@@ -21,16 +21,10 @@
     "typescript",
     "knowledge-management",
     "image-generation",
-    "playwright",
+    "agent-browser",
     "browser-automation"
   ],
   "mcpServers": {
-    "pw": {
-      "type": "stdio",
-      "command": "npx",
-      "args": ["-y", "@playwright/mcp@latest"],
-      "env": {}
-    },
     "context7": {
       "type": "http",
       "url": "https://mcp.context7.com/mcp"
diff --git a/plugins/compound-engineering/README.md b/plugins/compound-engineering/README.md
index 1328545..d941125 100644
--- a/plugins/compound-engineering/README.md
+++ b/plugins/compound-engineering/README.md
@@ -8,8 +8,8 @@ AI-powered development tools that get smarter with every use. Make each unit of
 |-----------|-------|
 | Agents | 27 |
 | Commands | 20 |
-| Skills | 13 |
-| MCP Servers | 2 |
+| Skills | 14 |
+| MCP Servers | 1 |
 
 ## Agents
 
@@ -96,7 +96,7 @@ Core workflow commands use `workflows:` prefix to avoid collisions with built-in
 | `/resolve_pr_parallel` | Resolve PR comments in parallel |
 | `/resolve_todo_parallel` | Resolve todos in parallel |
 | `/triage` | Triage and prioritize issues |
-| `/playwright-test` | Run browser tests on PR-affected pages |
+| `/test-browser` | Run browser tests on PR-affected pages |
 | `/xcode-test` | Build and test iOS apps on simulator |
 | `/feature-video` | Record video walkthroughs and add to PR description |
 
@@ -134,6 +134,12 @@ Core workflow commands use `workflows:` prefix to avoid collisions with built-in
 |-------|-------------|
 | `rclone` | Upload files to S3, Cloudflare R2, Backblaze B2, and cloud storage |
 
+### Browser Automation
+
+| Skill | Description |
+|-------|-------------|
+| `agent-browser` | CLI-based browser automation using Vercel's agent-browser |
+
 ### Image Generation
 
 | Skill | Description |
@@ -154,19 +160,8 @@ Core workflow commands use `workflows:` prefix to avoid collisions with built-in
 
 | Server | Description |
 |--------|-------------|
-| `playwright` | Browser automation via `@playwright/mcp` |
 | `context7` | Framework documentation lookup via Context7 |
 
-### Playwright
-
-**Tools provided:**
-- `browser_navigate` - Navigate to URLs
-- `browser_take_screenshot` - Take screenshots
-- `browser_click` - Click elements
-- `browser_fill_form` - Fill form fields
-- `browser_snapshot` - Get accessibility snapshot
-- `browser_evaluate` - Execute JavaScript
-
 ### Context7
 
 **Tools provided:**
@@ -177,6 +172,17 @@ Supports 100+ frameworks including Rails, React, Next.js, Vue, Django, Laravel,
 
 MCP servers start automatically when the plugin is enabled.
 
+## Browser Automation
+
+This plugin uses **agent-browser CLI** for browser automation tasks. Install it globally:
+
+```bash
+npm install -g agent-browser
+agent-browser install  # Downloads Chromium
+```
+
+The `agent-browser` skill provides comprehensive documentation on usage.
+
 ## Installation
 
 ```bash
@@ -187,19 +193,13 @@ claude /plugin install compound-engineering
 
 ### MCP Servers Not Auto-Loading
 
-**Issue:** The bundled MCP servers (Playwright and Context7) may not load automatically when the plugin is installed.
+**Issue:** The bundled Context7 MCP server may not load automatically when the plugin is installed.
 
-**Workaround:** Manually add them to your project's `.claude/settings.json`:
+**Workaround:** Manually add it to your project's `.claude/settings.json`:
 
 ```json
 {
   "mcpServers": {
-    "playwright": {
-      "type": "stdio",
-      "command": "npx",
-      "args": ["-y", "@playwright/mcp@latest"],
-      "env": {}
-    },
     "context7": {
       "type": "http",
       "url": "https://mcp.context7.com/mcp"
@@ -208,7 +208,7 @@ claude /plugin install compound-engineering
 }
 ```
 
-Or add them globally in `~/.claude/settings.json` for all projects.
+Or add it globally in `~/.claude/settings.json` for all projects.
 
 ## Version History
 
diff --git a/plugins/compound-engineering/agents/design/design-implementation-reviewer.md b/plugins/compound-engineering/agents/design/design-implementation-reviewer.md
index a867b41..0debf3c 100644
--- a/plugins/compound-engineering/agents/design/design-implementation-reviewer.md
+++ b/plugins/compound-engineering/agents/design/design-implementation-reviewer.md
@@ -11,11 +11,20 @@ Your primary responsibility is to conduct thorough visual comparisons between im
 ## Your Workflow
 
 1. **Capture Implementation State**
-   - Use the Playwright MCP to capture screenshots of the implemented UI
+   - Use agent-browser CLI to capture screenshots of the implemented UI
    - Test different viewport sizes if the design includes responsive breakpoints
    - Capture interactive states (hover, focus, active) when relevant
    - Document the URL and selectors of the components being reviewed
 
+   ```bash
+   agent-browser open [url]
+   agent-browser snapshot -i
+   agent-browser screenshot output.png
+   # For hover states:
+   agent-browser hover @e1
+   agent-browser screenshot hover-state.png
+   ```
+
 2. **Retrieve Design Specifications**
    - Use the Figma MCP to access the corresponding design files
    - Extract design tokens (colors, typography, spacing, shadows)
diff --git a/plugins/compound-engineering/agents/design/design-iterator.md b/plugins/compound-engineering/agents/design/design-iterator.md
index b266df6..34f2c65 100644
--- a/plugins/compound-engineering/agents/design/design-iterator.md
+++ b/plugins/compound-engineering/agents/design/design-iterator.md
@@ -23,50 +23,42 @@ For each iteration cycle, you must:
 
 ### Setup: Set Appropriate Window Size
 
-Before starting iterations, resize the browser to fit your target area:
+Before starting iterations, open the browser in headed mode to see and resize as needed:
 
+```bash
+agent-browser --headed open [url]
 ```
-browser_resize with width and height appropriate for the component:
+
+Recommended viewport sizes for reference:
 - Small component (button, card): 800x600
 - Medium section (hero, features): 1200x800
 - Full page section: 1440x900
-```
 
 ### Taking Element Screenshots
 
-Use `browser_take_screenshot` with element targeting:
+1. First, get element references with `agent-browser snapshot -i`
+2. Find the ref for your target element (e.g., @e1, @e2)
+3. Use `agent-browser scrollintoview @e1` to focus on specific elements
+4. Take screenshot: `agent-browser screenshot output.png`
 
-1. First, take a `browser_snapshot` to get element references
-2. Find the `ref` for your target element (e.g., a section, div, or component)
-3. Screenshot that specific element:
+### Viewport Screenshots
 
-```
-browser_take_screenshot with:
-- element: "Hero section" (human-readable description)
-- ref: "E123" (exact ref from snapshot)
-```
-
-### Fallback: Viewport Screenshots
-
-If the element doesn't have a clear ref, ensure the browser viewport shows only your target area:
-
-1. Use `browser_resize` to set viewport to component dimensions
-2. Scroll the element into view using `browser_evaluate`
-3. Take a viewport screenshot (no element/ref params)
+For focused screenshots:
+1. Use `agent-browser scrollintoview @e1` to scroll element into view
+2. Take viewport screenshot: `agent-browser screenshot output.png`
 
 ### Example Workflow
 
-```
-1. browser_resize(width: 1200, height: 800)
-2. browser_navigate to page
-3. browser_snapshot to see element refs
-4. browser_take_screenshot(element: "Features grid", ref: "E45")
-5. [analyze and implement changes]
-6. browser_take_screenshot(element: "Features grid", ref: "E45")
-7. [repeat...]
+```bash
+1. agent-browser open [url]
+2. agent-browser snapshot -i  # Get refs
+3. agent-browser screenshot output.png
+4. [analyze and implement changes]
+5. agent-browser screenshot output-v2.png
+6. [repeat...]
 ```
 
-**Never use `fullPage: true`** - it captures unnecessary content and bloats context.
+**Keep screenshots focused** - capture only the element/area you're working on to reduce noise.
 
 ## Design Principles to Apply
 
diff --git a/plugins/compound-engineering/agents/design/figma-design-sync.md b/plugins/compound-engineering/agents/design/figma-design-sync.md
index 1b30180..bb58632 100644
--- a/plugins/compound-engineering/agents/design/figma-design-sync.md
+++ b/plugins/compound-engineering/agents/design/figma-design-sync.md
@@ -11,7 +11,13 @@ You are an expert design-to-code synchronization specialist with deep expertise
 
 1. **Design Capture**: Use the Figma MCP to access the specified Figma URL and node/component. Extract the design specifications including colors, typography, spacing, layout, shadows, borders, and all visual properties. Also take a screenshot and load it into the agent.
 
-2. **Implementation Capture**: Use the Playwright MCP to navigate to the specified web page/component URL and capture a high-quality screenshot of the current implementation.
+2. **Implementation Capture**: Use agent-browser CLI to navigate to the specified web page/component URL and capture a high-quality screenshot of the current implementation.
+
+   ```bash
+   agent-browser open [url]
+   agent-browser snapshot -i
+   agent-browser screenshot implementation.png
+   ```
 
 3. **Systematic Comparison**: Perform a meticulous visual comparison between the Figma design and the screenshot, analyzing:
 
diff --git a/plugins/compound-engineering/agents/workflow/bug-reproduction-validator.md b/plugins/compound-engineering/agents/workflow/bug-reproduction-validator.md
index 0b1f510..555ba4b 100644
--- a/plugins/compound-engineering/agents/workflow/bug-reproduction-validator.md
+++ b/plugins/compound-engineering/agents/workflow/bug-reproduction-validator.md
@@ -19,7 +19,7 @@ When presented with a bug report, you will:
    - Set up the minimal test case needed to reproduce the issue
    - Execute the reproduction steps methodically, documenting each step
    - If the bug involves data states, check fixtures or create appropriate test data
-   - For UI bugs, consider using Playwright MCP if available to visually verify
+   - For UI bugs, use agent-browser CLI to visually verify (see `agent-browser` skill)
    - For backend bugs, examine logs, database states, and service interactions
 
 3. **Validation Methodology**:
diff --git a/plugins/compound-engineering/commands/feature-video.md b/plugins/compound-engineering/commands/feature-video.md
index 65c104b..346f765 100644
--- a/plugins/compound-engineering/commands/feature-video.md
+++ b/plugins/compound-engineering/commands/feature-video.md
@@ -13,7 +13,7 @@ argument-hint: "[PR number or 'current'] [optional: base URL, default localhost:
 <role>Developer Relations Engineer creating feature demo videos</role>
 
 This command creates professional video walkthroughs of features for PR documentation:
-- Records browser interactions using Playwright video capture
+- Records browser interactions using agent-browser CLI
 - Demonstrates the complete user flow
 - Uploads the video for easy sharing
 - Updates the PR description with an embedded video
@@ -22,12 +22,26 @@ This command creates professional video walkthroughs of features for PR document
 
 <requirements>
 - Local development server running (e.g., `bin/dev`, `rails server`)
-- Playwright MCP server connected
+- agent-browser CLI installed
 - Git repository with a PR to document
 - `ffmpeg` installed (for video conversion)
 - `rclone` configured (optional, for cloud upload - see rclone skill)
 </requirements>
 
+## Setup
+
+**Check installation:**
+```bash
+command -v agent-browser >/dev/null 2>&1 && echo "Installed" || echo "NOT INSTALLED"
+```
+
+**Install if needed:**
+```bash
+npm install -g agent-browser && agent-browser install
+```
+
+See the `agent-browser` skill for detailed usage.
+
 ## Main Tasks
 
 ### 1. Parse Arguments
@@ -118,26 +132,9 @@ Does this look right?
 mkdir -p tmp/videos
 ```
 
-**Start browser with video recording using Playwright MCP:**
+**Recording approach: Use browser screenshots as frames**
 
-Note: Playwright MCP's browser_navigate will be used, and we'll use browser_run_code to enable video recording:
-
-```javascript
-// Enable video recording context
-mcp__plugin_compound-engineering_pw__browser_run_code({
-  code: `async (page) => {
-    // Video recording is enabled at context level
-    // The MCP server handles this automatically
-    return 'Video recording active';
-  }`
-})
-```
-
-**Alternative: Use browser screenshots as frames**
-
-If video recording isn't available via MCP, fall back to:
-1. Take screenshots at key moments
-2. Combine into a GIF using ffmpeg
+agent-browser captures screenshots at key moments, then combine into video using ffmpeg:
 
 ```bash
 ffmpeg -framerate 2 -pattern_type glob -i 'tmp/screenshots/*.png' -vf "scale=1280:-1" tmp/videos/feature-demo.gif
@@ -152,32 +149,32 @@ ffmpeg -framerate 2 -pattern_type glob -i 'tmp/screenshots/*.png' -vf "scale=128
 Execute the planned flow, capturing each step:
 
 **Step 1: Navigate to starting point**
-```
-mcp__plugin_compound-engineering_pw__browser_navigate({ url: "[base-url]/[start-route]" })
-mcp__plugin_compound-engineering_pw__browser_wait_for({ time: 2 })
-mcp__plugin_compound-engineering_pw__browser_take_screenshot({ filename: "tmp/screenshots/01-start.png" })
+```bash
+agent-browser open "[base-url]/[start-route]"
+agent-browser wait 2000
+agent-browser screenshot tmp/screenshots/01-start.png
 ```
 
 **Step 2: Perform navigation/interactions**
-```
-mcp__plugin_compound-engineering_pw__browser_click({ element: "[description]", ref: "[ref]" })
-mcp__plugin_compound-engineering_pw__browser_wait_for({ time: 1 })
-mcp__plugin_compound-engineering_pw__browser_take_screenshot({ filename: "tmp/screenshots/02-navigate.png" })
+```bash
+agent-browser snapshot -i  # Get refs
+agent-browser click @e1    # Click navigation element
+agent-browser wait 1000
+agent-browser screenshot tmp/screenshots/02-navigate.png
 ```
 
 **Step 3: Demonstrate feature**
-```
-mcp__plugin_compound-engineering_pw__browser_snapshot({})
-// Identify interactive elements
-mcp__plugin_compound-engineering_pw__browser_click({ element: "[feature element]", ref: "[ref]" })
-mcp__plugin_compound-engineering_pw__browser_wait_for({ time: 1 })
-mcp__plugin_compound-engineering_pw__browser_take_screenshot({ filename: "tmp/screenshots/03-feature.png" })
+```bash
+agent-browser snapshot -i  # Get refs for feature elements
+agent-browser click @e2    # Click feature element
+agent-browser wait 1000
+agent-browser screenshot tmp/screenshots/03-feature.png
 ```
 
 **Step 4: Capture result**
-```
-mcp__plugin_compound-engineering_pw__browser_wait_for({ time: 2 })
-mcp__plugin_compound-engineering_pw__browser_take_screenshot({ filename: "tmp/screenshots/04-result.png" })
+```bash
+agent-browser wait 2000
+agent-browser screenshot tmp/screenshots/04-result.png
 ```
 
 **Create video/GIF from screenshots:**
@@ -189,17 +186,14 @@ mkdir -p tmp/videos tmp/screenshots
 # Create MP4 video (RECOMMENDED - better quality, smaller size)
 # -framerate 0.5 = 2 seconds per frame (slower playback)
 # -framerate 1 = 1 second per frame
-ffmpeg -y -framerate 0.5 -pattern_type glob -i '.playwright-mcp/tmp/screenshots/*.png' \
+ffmpeg -y -framerate 0.5 -pattern_type glob -i 'tmp/screenshots/*.png' \
   -c:v libx264 -pix_fmt yuv420p -vf "scale=1280:-2" \
   tmp/videos/feature-demo.mp4
 
 # Create low-quality GIF for preview (small file, for GitHub embed)
-ffmpeg -y -framerate 0.5 -pattern_type glob -i '.playwright-mcp/tmp/screenshots/*.png' \
+ffmpeg -y -framerate 0.5 -pattern_type glob -i 'tmp/screenshots/*.png' \
   -vf "scale=640:-1:flags=lanczos,split[s0][s1];[s0]palettegen=max_colors=128[p];[s1][p]paletteuse" \
   -loop 0 tmp/videos/feature-demo-preview.gif
-
-# Copy screenshots to project folder for easy access
-cp -r .playwright-mcp/tmp/screenshots tmp/
 ```
 
 **Note:**
diff --git a/plugins/compound-engineering/commands/playwright-test.md b/plugins/compound-engineering/commands/test-browser.md
similarity index 69%
rename from plugins/compound-engineering/commands/playwright-test.md
rename to plugins/compound-engineering/commands/test-browser.md
index 44472d1..9ee9658 100644
--- a/plugins/compound-engineering/commands/playwright-test.md
+++ b/plugins/compound-engineering/commands/test-browser.md
@@ -1,12 +1,12 @@
 ---
-name: playwright-test
-description: Run Playwright browser tests on pages affected by current PR or branch
+name: test-browser
+description: Run browser tests on pages affected by current PR or branch
 argument-hint: "[PR number, branch name, or 'current' for current branch]"
 ---
 
-# Playwright Test Command
+# Browser Test Command
 
-<command_purpose>Run end-to-end browser tests on pages affected by a PR or branch changes using Playwright MCP.</command_purpose>
+<command_purpose>Run end-to-end browser tests on pages affected by a PR or branch changes using agent-browser CLI.</command_purpose>
 
 ## Introduction
 
@@ -22,10 +22,25 @@ This command tests affected pages in a real browser, catching issues that unit t
 
 <requirements>
 - Local development server running (e.g., `bin/dev`, `rails server`)
-- Playwright MCP server connected
+- agent-browser CLI installed
 - Git repository with changes to test
 </requirements>
 
+## Setup
+
+**Check installation:**
+```bash
+command -v agent-browser >/dev/null 2>&1 && echo "Installed" || echo "NOT INSTALLED"
+```
+
+**Install if needed:**
+```bash
+npm install -g agent-browser
+agent-browser install  # Downloads Chromium
+```
+
+See the `agent-browser` skill for detailed usage.
+
 ## Main Tasks
 
 ### 1. Determine Test Scope
@@ -77,9 +92,9 @@ Build a list of URLs to test based on the mapping.
 
 Before testing, verify the local server is accessible:
 
-```
-mcp__playwright__browser_navigate({ url: "http://localhost:3000" })
-mcp__playwright__browser_snapshot({})
+```bash
+agent-browser open http://localhost:3000
+agent-browser snapshot -i
 ```
 
 If server is not running, inform user:
@@ -90,7 +105,7 @@ Please start your development server:
 - Rails: `bin/dev` or `rails server`
 - Node: `npm run dev`
 
-Then run `/playwright-test` again.
+Then run `/test-browser` again.
 ```
 
 </check_server>
@@ -102,26 +117,27 @@ Then run `/playwright-test` again.
 For each affected route:
 
 **Step 1: Navigate and capture snapshot**
-```
-mcp__playwright__browser_navigate({ url: "http://localhost:3000/[route]" })
-mcp__playwright__browser_snapshot({})
+```bash
+agent-browser open "http://localhost:3000/[route]"
+agent-browser snapshot -i
 ```
 
-**Step 2: Check for errors**
-```
-mcp__playwright__browser_console_messages({ level: "error" })
+**Step 2: Check for errors** (use headed mode for console inspection)
+```bash
+agent-browser --headed open "http://localhost:3000/[route]"
 ```
 
 **Step 3: Verify key elements**
+- Use `agent-browser snapshot -i` to get interactive elements with refs
 - Page title/heading present
 - Primary content rendered
 - No error messages visible
 - Forms have expected fields
 
-**Step 4: Test critical interactions (if applicable)**
-```
-mcp__playwright__browser_click({ element: "[description]", ref: "[ref]" })
-mcp__playwright__browser_snapshot({})
+**Step 4: Test critical interactions**
+```bash
+agent-browser click @e1  # Use ref from snapshot
+agent-browser snapshot -i
 ```
 
 </test_pages>
@@ -162,8 +178,7 @@ Did it work correctly?
 When a test fails:
 
 1. **Document the failure:**
-   - Screenshot the error state
-   - Capture console errors
+   - Screenshot the error state: `agent-browser screenshot error.png`
    - Note the exact reproduction steps
 
 2. **Ask user how to proceed:**
@@ -186,7 +201,7 @@ When a test fails:
    - Re-run the failing test
 
 4. **If "Create todo":**
-   - Create `{id}-pending-p1-playwright-{description}.md`
+   - Create `{id}-pending-p1-browser-test-{description}.md`
    - Continue testing
 
 5. **If "Skip":**
@@ -202,7 +217,7 @@ When a test fails:
 After all tests complete, present summary:
 
 ```markdown
-## 🎭 Playwright Test Results
+## Browser Test Results
 
 **Test Scope:** PR #[number] / [branch name]
 **Server:** http://localhost:3000
@@ -211,23 +226,23 @@ After all tests complete, present summary:
 
 | Route | Status | Notes |
 |-------|--------|-------|
-| `/users` | ✅ Pass | |
-| `/settings` | ✅ Pass | |
-| `/dashboard` | ❌ Fail | Console error: [msg] |
-| `/checkout` | ⏭️ Skip | Requires payment credentials |
+| `/users` | Pass | |
+| `/settings` | Pass | |
+| `/dashboard` | Fail | Console error: [msg] |
+| `/checkout` | Skip | Requires payment credentials |
 
 ### Console Errors: [count]
 - [List any errors found]
 
 ### Human Verifications: [count]
-- OAuth flow: ✅ Confirmed
-- Email delivery: ✅ Confirmed
+- OAuth flow: Confirmed
+- Email delivery: Confirmed
 
 ### Failures: [count]
 - `/dashboard` - [issue description]
 
 ### Created Todos: [count]
-- `005-pending-p1-playwright-dashboard-error.md`
+- `005-pending-p1-browser-test-dashboard-error.md`
 
 ### Result: [PASS / FAIL / PARTIAL]
 ```
@@ -238,11 +253,22 @@ After all tests complete, present summary:
 
 ```bash
 # Test current branch changes
-/playwright-test
+/test-browser
 
 # Test specific PR
-/playwright-test 847
+/test-browser 847
 
 # Test specific branch
-/playwright-test feature/new-dashboard
+/test-browser feature/new-dashboard
+```
+
+## Key agent-browser Commands
+
+```bash
+agent-browser open <url>           # Navigate
+agent-browser snapshot -i          # Interactive elements with refs
+agent-browser click @e1            # Click by ref
+agent-browser fill @e1 "text"      # Fill input
+agent-browser screenshot out.png   # Screenshot
+agent-browser --headed open <url>  # Visible browser
 ```
diff --git a/plugins/compound-engineering/commands/workflows/review.md b/plugins/compound-engineering/commands/workflows/review.md
index 9d26516..7431c62 100644
--- a/plugins/compound-engineering/commands/workflows/review.md
+++ b/plugins/compound-engineering/commands/workflows/review.md
@@ -445,8 +445,8 @@ After presenting the Summary Report, offer appropriate testing based on project
 
 **For Web Projects:**
 ```markdown
-**"Want to run Playwright browser tests on the affected pages?"**
-1. Yes - run `/playwright-test`
+**"Want to run browser tests on the affected pages?"**
+1. Yes - run `/test-browser`
 2. No - skip
 ```
 
@@ -460,7 +460,7 @@ After presenting the Summary Report, offer appropriate testing based on project
 **For Hybrid Projects (e.g., Rails + Hotwire Native):**
 ```markdown
 **"Want to run end-to-end tests?"**
-1. Web only - run `/playwright-test`
+1. Web only - run `/test-browser`
 2. iOS only - run `/xcode-test`
 3. Both - run both commands
 4. No - skip
@@ -470,22 +470,22 @@ After presenting the Summary Report, offer appropriate testing based on project
 
 #### If User Accepts Web Testing:
 
-Spawn a subagent to run Playwright tests (preserves main context):
+Spawn a subagent to run browser tests (preserves main context):
 
 ```
-Task general-purpose("Run /playwright-test for PR #[number]. Test all affected pages, check for console errors, handle failures by creating todos and fixing.")
+Task general-purpose("Run /test-browser for PR #[number]. Test all affected pages, check for console errors, handle failures by creating todos and fixing.")
 ```
 
 The subagent will:
 1. Identify pages affected by the PR
-2. Navigate to each page and capture snapshots
+2. Navigate to each page and capture snapshots (using Playwright MCP or agent-browser CLI)
 3. Check for console errors
 4. Test critical interactions
 5. Pause for human verification on OAuth/email/payment flows
 6. Create P1 todos for any failures
 7. Fix and retry until all tests pass
 
-**Standalone:** `/playwright-test [PR number]`
+**Standalone:** `/test-browser [PR number]`
 
 #### If User Accepts iOS Testing:
 
diff --git a/plugins/compound-engineering/commands/workflows/work.md b/plugins/compound-engineering/commands/workflows/work.md
index bfa5dfd..328fb78 100644
--- a/plugins/compound-engineering/commands/workflows/work.md
+++ b/plugins/compound-engineering/commands/workflows/work.md
@@ -181,11 +181,13 @@ This command takes a work document (plan, specification, or todo file) and execu
    bin/dev  # Run in background
    ```
 
-   **Step 2: Capture screenshots with Playwright MCP tools**
-   - `browser_navigate` to go to affected pages
-   - `browser_resize` to set viewport (desktop or mobile as needed)
-   - `browser_snapshot` to verify page state
-   - `browser_take_screenshot` to capture images
+   **Step 2: Capture screenshots with agent-browser CLI**
+   ```bash
+   agent-browser open http://localhost:3000/[route]
+   agent-browser snapshot -i
+   agent-browser screenshot output.png
+   ```
+   See the `agent-browser` skill for detailed usage.
 
    **Step 3: Upload using imgup skill**
    ```bash
diff --git a/plugins/compound-engineering/skills/agent-browser/SKILL.md b/plugins/compound-engineering/skills/agent-browser/SKILL.md
new file mode 100644
index 0000000..3ff264e
--- /dev/null
+++ b/plugins/compound-engineering/skills/agent-browser/SKILL.md
@@ -0,0 +1,223 @@
+---
+name: agent-browser
+description: Browser automation using Vercel's agent-browser CLI. Use when you need to interact with web pages, fill forms, take screenshots, or scrape data. Alternative to Playwright MCP - uses Bash commands with ref-based element selection. Triggers on "browse website", "fill form", "click button", "take screenshot", "scrape page", "web automation".
+---
+
+# agent-browser: CLI Browser Automation
+
+Vercel's headless browser automation CLI designed for AI agents. Uses ref-based selection (@e1, @e2) from accessibility snapshots.
+
+## Setup Check
+
+```bash
+# Check installation
+command -v agent-browser >/dev/null 2>&1 && echo "Installed" || echo "NOT INSTALLED - run: npm install -g agent-browser && agent-browser install"
+```
+
+### Install if needed
+
+```bash
+npm install -g agent-browser
+agent-browser install  # Downloads Chromium
+```
+
+## Core Workflow
+
+**The snapshot + ref pattern is optimal for LLMs:**
+
+1. **Navigate** to URL
+2. **Snapshot** to get interactive elements with refs
+3. **Interact** using refs (@e1, @e2, etc.)
+4. **Re-snapshot** after navigation or DOM changes
+
+```bash
+# Step 1: Open URL
+agent-browser open https://example.com
+
+# Step 2: Get interactive elements with refs
+agent-browser snapshot -i --json
+
+# Step 3: Interact using refs
+agent-browser click @e1
+agent-browser fill @e2 "search query"
+
+# Step 4: Re-snapshot after changes
+agent-browser snapshot -i
+```
+
+## Key Commands
+
+### Navigation
+
+```bash
+agent-browser open <url>       # Navigate to URL
+agent-browser back             # Go back
+agent-browser forward          # Go forward
+agent-browser reload           # Reload page
+agent-browser close            # Close browser
+```
+
+### Snapshots (Essential for AI)
+
+```bash
+agent-browser snapshot              # Full accessibility tree
+agent-browser snapshot -i           # Interactive elements only (recommended)
+agent-browser snapshot -i --json    # JSON output for parsing
+agent-browser snapshot -c           # Compact (remove empty elements)
+agent-browser snapshot -d 3         # Limit depth
+```
+
+### Interactions
+
+```bash
+agent-browser click @e1                    # Click element
+agent-browser dblclick @e1                 # Double-click
+agent-browser fill @e1 "text"              # Clear and fill input
+agent-browser type @e1 "text"              # Type without clearing
+agent-browser press Enter                  # Press key
+agent-browser hover @e1                    # Hover element
+agent-browser check @e1                    # Check checkbox
+agent-browser uncheck @e1                  # Uncheck checkbox
+agent-browser select @e1 "option"          # Select dropdown option
+agent-browser scroll down 500              # Scroll (up/down/left/right)
+agent-browser scrollintoview @e1           # Scroll element into view
+```
+
+### Get Information
+
+```bash
+agent-browser get text @e1          # Get element text
+agent-browser get html @e1          # Get element HTML
+agent-browser get value @e1         # Get input value
+agent-browser get attr href @e1     # Get attribute
+agent-browser get title             # Get page title
+agent-browser get url               # Get current URL
+agent-browser get count "button"    # Count matching elements
+```
+
+### Screenshots & PDFs
+
+```bash
+agent-browser screenshot                      # Viewport screenshot
+agent-browser screenshot --full               # Full page
+agent-browser screenshot output.png           # Save to file
+agent-browser screenshot --full output.png    # Full page to file
+agent-browser pdf output.pdf                  # Save as PDF
+```
+
+### Wait
+
+```bash
+agent-browser wait @e1              # Wait for element
+agent-browser wait 2000             # Wait milliseconds
+agent-browser wait "text"           # Wait for text to appear
+```
+
+## Semantic Locators (Alternative to Refs)
+
+```bash
+agent-browser find role button click --name "Submit"
+agent-browser find text "Sign up" click
+agent-browser find label "Email" fill "user@example.com"
+agent-browser find placeholder "Search..." fill "query"
+```
+
+## Sessions (Parallel Browsers)
+
+```bash
+# Run multiple independent browser sessions
+agent-browser --session browser1 open https://site1.com
+agent-browser --session browser2 open https://site2.com
+
+# List active sessions
+agent-browser session list
+```
+
+## Examples
+
+### Login Flow
+
+```bash
+agent-browser open https://app.example.com/login
+agent-browser snapshot -i
+# Output shows: textbox "Email" [ref=e1], textbox "Password" [ref=e2], button "Sign in" [ref=e3]
+agent-browser fill @e1 "user@example.com"
+agent-browser fill @e2 "password123"
+agent-browser click @e3
+agent-browser wait 2000
+agent-browser snapshot -i  # Verify logged in
+```
+
+### Search and Extract
+
+```bash
+agent-browser open https://news.ycombinator.com
+agent-browser snapshot -i --json
+# Parse JSON to find story links
+agent-browser get text @e12  # Get headline text
+agent-browser click @e12     # Click to open story
+```
+
+### Form Filling
+
+```bash
+agent-browser open https://forms.example.com
+agent-browser snapshot -i
+agent-browser fill @e1 "John Doe"
+agent-browser fill @e2 "john@example.com"
+agent-browser select @e3 "United States"
+agent-browser check @e4  # Agree to terms
+agent-browser click @e5  # Submit button
+agent-browser screenshot confirmation.png
+```
+
+### Debug Mode
+
+```bash
+# Run with visible browser window
+agent-browser --headed open https://example.com
+agent-browser --headed snapshot -i
+agent-browser --headed click @e1
+```
+
+## JSON Output
+
+Add `--json` for structured output:
+
+```bash
+agent-browser snapshot -i --json
+```
+
+Returns:
+```json
+{
+  "success": true,
+  "data": {
+    "refs": {
+      "e1": {"name": "Submit", "role": "button"},
+      "e2": {"name": "Email", "role": "textbox"}
+    },
+    "snapshot": "- button \"Submit\" [ref=e1]\n- textbox \"Email\" [ref=e2]"
+  }
+}
+```
+
+## vs Playwright MCP
+
+| Feature | agent-browser (CLI) | Playwright MCP |
+|---------|---------------------|----------------|
+| Interface | Bash commands | MCP tools |
+| Selection | Refs (@e1) | Refs (e1) |
+| Output | Text/JSON | Tool responses |
+| Parallel | Sessions | Tabs |
+| Best for | Quick automation | Tool integration |
+
+Use agent-browser when:
+- You prefer Bash-based workflows
+- You want simpler CLI commands
+- You need quick one-off automation
+
+Use Playwright MCP when:
+- You need deep MCP tool integration
+- You want tool-based responses
+- You're building complex automation