Advanced samples (#16)

- Moves persisted browser sample to sample-app - also just creates a persisted browser (I find I often need one for manual testing)
- advanced-samples actions: example CAPTCHA solver

Author: juecd

README.md

@@ -38,7 +38,7 @@ create-kernel-app [app-name] [options]
   - `sample-app`: Basic template with Playwright integration
   - `browser-use`: Template with Browser Use SDK (Python only)
   - `stagehand`: Template with Stagehand SDK (Typescript only)
-  - `persistent-browser`: Implements `sample-app` using a persistent browser 
+  - `advanced-sample`: Implements sample apps using advanced Kernel configs
   - `computer-use`: Implements a prompt loop using Anthropic Computer Use
 
 ### Examples
@@ -125,7 +125,7 @@ These are the sample apps currently available when you run `npx @onkernel/create
 | **sample-app** | Returns the page title of a specified URL | Playwright | `{ url }` |
 | **browser-use** | Completes a specified task | Browser Use | `{ task }` |
 | **stagehand** | Returns the first result of a specified Google search | Stagehand | `{ query }` |
-| **persistent-browser** | Implements `sample-app` using a persistent browser | Playwright | `{ url }` |
+| **advanced-sample** | Implements sample apps using advanced Kernel configs | n/a |
 | **computer-use** | Implements a prompt loop | Anthropic Computer Use API | `{ query }` |
 
 ## Documentation

index.ts

@@ -17,7 +17,7 @@ type TemplateKey =
   | "sample-app"
   | "browser-use"
   | "stagehand"
-  | "persistent-browser"
+  | "advanced-sample"
   | "computer-use";
 type LanguageInfo = { name: string; shorthand: string };
 type TemplateInfo = {
@@ -32,7 +32,7 @@ const LANGUAGE_PYTHON = "python";
 const TEMPLATE_SAMPLE_APP = "sample-app";
 const TEMPLATE_BROWSER_USE = "browser-use";
 const TEMPLATE_STAGEHAND = "stagehand";
-const TEMPLATE_PERSISTENT_BROWSER = "persistent-browser";
+const TEMPLATE_ADVANCED_SAMPLE = "advanced-sample";
 const TEMPLATE_COMPUTER_USE = "computer-use";
 const LANGUAGE_SHORTHAND_TS = "ts";
 const LANGUAGE_SHORTHAND_PY = "py";
@@ -62,10 +62,10 @@ const TEMPLATES: Record<TemplateKey, TemplateInfo> = {
     description: "Implements the Stagehand SDK",
     languages: [LANGUAGE_TYPESCRIPT],
   },
-  [TEMPLATE_PERSISTENT_BROWSER]: {
-    name: "Persistent Browser",
+  [TEMPLATE_ADVANCED_SAMPLE]: {
+    name: "Advanced Samples",
     description:
-      "Implements a persistent browser that maintains state across invocations",
+      "Implements sample actions with advanced Kernel configs",
     languages: [LANGUAGE_TYPESCRIPT, LANGUAGE_PYTHON],
   },
   [TEMPLATE_COMPUTER_USE]: {
@@ -84,8 +84,8 @@ const INVOKE_SAMPLES: Record<
       'kernel invoke ts-basic get-page-title --payload \'{"url": "https://www.google.com"}\'',
     [TEMPLATE_STAGEHAND]:
       'kernel invoke ts-stagehand stagehand-task --payload \'{"query": "Best wired earbuds"}\'',
-    [TEMPLATE_PERSISTENT_BROWSER]:
-      'kernel invoke ts-persistent-browser persistent-browser-task --payload \'{"url": "https://news.ycombinator.com/"}\'',
+    [TEMPLATE_ADVANCED_SAMPLE]:
+      'kernel invoke ts-advanced test-captcha-solver',
     [TEMPLATE_COMPUTER_USE]:
       'kernel invoke ts-cu cu-task --payload \'{"query": "Return the first url of a search result for NYC restaurant reviews Pete Wells"}\'',
   },
@@ -94,8 +94,8 @@ const INVOKE_SAMPLES: Record<
       'kernel invoke python-basic get-page-title --payload \'{"url": "https://www.google.com"}\'',
     [TEMPLATE_BROWSER_USE]:
       'kernel invoke python-bu bu-task --payload \'{"task": "Compare the price of gpt-4o and DeepSeek-V3"}\'',
-    [TEMPLATE_PERSISTENT_BROWSER]:
-      'kernel invoke python-persistent-browser persistent-browser-task --payload \'{"url": "https://news.ycombinator.com/"}\'',
+    [TEMPLATE_ADVANCED_SAMPLE]:
+      'kernel invoke python-advanced test-captcha-solver',
     [TEMPLATE_COMPUTER_USE]:
       'kernel invoke python-cu cu-task --payload \'{"query": "Return the first url of a search result for NYC restaurant reviews Pete Wells"}\'',
   },
@@ -110,8 +110,8 @@ const REGISTERED_APP_NAMES: Record<
       'ts-basic',
     [TEMPLATE_STAGEHAND]:
       'ts-stagehand',
-    [TEMPLATE_PERSISTENT_BROWSER]:
-      'ts-persistent-browser',
+    [TEMPLATE_ADVANCED_SAMPLE]:
+      'ts-advanced',
     [TEMPLATE_COMPUTER_USE]:
       'ts-cu',
   },
@@ -120,8 +120,8 @@ const REGISTERED_APP_NAMES: Record<
       'python-basic',
     [TEMPLATE_BROWSER_USE]:
       'python-bu',
-    [TEMPLATE_PERSISTENT_BROWSER]:
-      'python-persistent-browser',
+    [TEMPLATE_ADVANCED_SAMPLE]:
+      'python-advanced',
     [TEMPLATE_COMPUTER_USE]:
       'python-cu',
   },
@@ -336,13 +336,13 @@ function printNextSteps(
 ): void {
   // Determine which sample command to show based on language and template
   const deployCommand =
-    language === LANGUAGE_TYPESCRIPT && (template === TEMPLATE_SAMPLE_APP || template === TEMPLATE_PERSISTENT_BROWSER)
+    language === LANGUAGE_TYPESCRIPT && (template === TEMPLATE_SAMPLE_APP || template === TEMPLATE_ADVANCED_SAMPLE)
       ? "kernel deploy index.ts"
       : language === LANGUAGE_TYPESCRIPT && template === TEMPLATE_STAGEHAND
       ? "kernel deploy index.ts --env OPENAI_API_KEY=XXX"
       : language === LANGUAGE_TYPESCRIPT && template === TEMPLATE_COMPUTER_USE
       ? "kernel deploy index.ts --env ANTHROPIC_API_KEY=XXX"
-      : language === LANGUAGE_PYTHON && template === TEMPLATE_SAMPLE_APP
+      : language === LANGUAGE_PYTHON && (template === TEMPLATE_SAMPLE_APP || template === TEMPLATE_ADVANCED_SAMPLE)
       ? "kernel deploy main.py"
       : language === LANGUAGE_PYTHON && template === TEMPLATE_BROWSER_USE
       ? "kernel deploy main.py --env OPENAI_API_KEY=XXX"
@@ -384,7 +384,7 @@ program
   )
   .option(
     "-t, --template <template>",
-    `Template type (${TEMPLATE_SAMPLE_APP}, ${TEMPLATE_BROWSER_USE}, ${TEMPLATE_STAGEHAND}, ${TEMPLATE_PERSISTENT_BROWSER}, ${TEMPLATE_COMPUTER_USE})`
+    `Template type (${TEMPLATE_SAMPLE_APP}, ${TEMPLATE_BROWSER_USE}, ${TEMPLATE_STAGEHAND}, ${TEMPLATE_ADVANCED_SAMPLE}, ${TEMPLATE_COMPUTER_USE})`
   )
   .action(
     async (

…tes/python/persistent-browser/.gitignore …plates/python/advanced-sample/.gitignoretemplates/python/persistent-browser/.gitignore renamed to templates/python/advanced-sample/.gitignore templates/python/persistent-browser/.gitignore renamed to templates/python/advanced-sample/.gitignore

@@ -0,0 +1,3 @@
+# Kernel Python Advanced Sample App
+
+This is a collection of Kernel actions that demonstrate how to use the Kernel SDK.

templates/python/advanced-sample/README.md

@@ -0,0 +1,54 @@
+import kernel
+from kernel import Kernel
+from playwright.async_api import async_playwright
+
+client = Kernel()
+app = kernel.App("python-advanced")
+
+"""
+Example showing Kernel's auto-CAPTCHA solver.
+Visit the live view url to see the Kernel browser auto-solve the CAPTCHA on the site.
+
+Args:
+    ctx: Kernel context containing invocation information
+Returns:
+    None
+    
+Invoke this via CLI:
+    export KERNEL_API_KEY=<your_api_key>
+    kernel deploy main.py  # If you haven't already deployed this app
+    kernel invoke py-advanced test_captcha_solver
+    kernel logs py-advanced -f  # Open in separate tab
+"""
+@app.action("test-captcha-solver")
+async def test_captcha_solver(ctx: kernel.KernelContext) -> None:
+    kernel_browser = client.browsers.create(
+        invocation_id=ctx.invocation_id,
+        stealth=True,
+        persistence={"id": "captcha-solver"}
+    )
+    
+    async with async_playwright() as playwright:
+        browser = await playwright.chromium.connect_over_cdp(kernel_browser.cdp_ws_url)
+        
+        # Get or create context and page
+        contexts = browser.contexts
+        context = contexts[0] if contexts else await browser.new_context()
+        pages = context.pages
+        page = pages[0] if pages else await context.new_page()
+        
+        # Access the live view. Retrieve this live_view_url from the Kernel logs in your CLI:
+        # export KERNEL_API_KEY=<Your API key>
+        # kernel logs py-advanced --follow
+        print("Kernel browser live view url: ", kernel_browser.browser_live_view_url)
+        
+        # Navigate to a site with a CAPTCHA
+        try:
+            await page.wait_for_timeout(10000)  # Add a delay to give you time to visit the live view url
+            await page.goto("https://www.google.com/recaptcha/api2/demo")
+        except Exception as e:
+            print(f"Error during navigation: {e}")
+            raise
+        # Watch Kernel auto-solve the CAPTCHA!
+        await browser.close()
+

templates/python/advanced-sample/main.py

@@ -0,0 +1,13 @@
+[project]
+name = "kernel-python-advanced"
+version = "0.1.0"
+description = "Sample application implementing advanced Kernel configs"
+readme = "README.md"
+requires-python = ">=3.11"
+dependencies = [
+    "kernel>=0.5.0",
+    "playwright>=1.52.0"
+]
+
+[dependency-groups]
+dev = ["mypy>=1.15.0"]