pi-harness-runtime 1.1.181 → 1.1.182

package/harness/e2e-commands.ts

@@ -0,0 +1,216 @@
+/**
+ * E2E Testing Commands
+ * 
+ * Provides /e2e command for running WebMCP-powered E2E tests.
+ * 
+ * Usage:
+ *   /e2e test --scenario test-register.md
+ *   /e2e run --url https://app.com --steps "fill email, click submit"
+ *   /e2e status
+ */
+
+import type { ExtensionAPI, ExtensionContext } from "@earendil-works/pi-coding-agent";
+
+interface E2ECommandArgs {
+  subcommand: string;
+  url?: string;
+  scenario?: string;
+  steps?: string;
+  name?: string;
+  screenshot?: boolean;
+}
+
+interface TestResult {
+  name: string;
+  passed: boolean;
+  error?: string;
+  durationMs: number;
+}
+
+/**
+ * Register E2E testing commands
+ */
+export function registerE2ECommands(pi: ExtensionAPI): void {
+  pi.registerCommand("e2e", {
+    description: "Run E2E tests using WebMCP protocol",
+    handler: async (args: string, ctx: ExtensionContext): Promise<void> => {
+      const parsed = parseArgs(args);
+      
+      switch (parsed.subcommand) {
+        case "test":
+        case "run":
+          return handleRunTest(ctx, parsed);
+          
+        case "status":
+          return handleStatus(ctx);
+          
+        case "help":
+        default:
+          return handleHelp(ctx);
+      }
+    },
+  });
+}
+
+/**
+ * Parse command arguments
+ */
+function parseArgs(args: string): E2ECommandArgs {
+  const parts = args.trim().split(/\s+/);
+  const result: E2ECommandArgs = { subcommand: parts[0] || "help" };
+  
+  for (let i = 1; i < parts.length; i++) {
+    const part = parts[i];
+    
+    if (part === "--url" || part === "-u") {
+      result.url = parts[++i];
+    } else if (part === "--scenario" || part === "-s") {
+      result.scenario = parts[++i];
+    } else if (part === "--steps") {
+      result.steps = parts.slice(i + 1).join(" ");
+      break;
+    } else if (part === "--name" || part === "-n") {
+      result.name = parts[++i];
+    } else if (part === "--screenshot") {
+      result.screenshot = true;
+    }
+  }
+  
+  return result;
+}
+
+/**
+ * Handle /e2e test or /e2e run
+ */
+async function handleRunTest(ctx: ExtensionContext, args: E2ECommandArgs): Promise<void> {
+  if (!args.url && !args.scenario) {
+    ctx.ui.notify(
+      "Usage: /e2e test --url <url> [--name <name>] [--steps <steps>] [--screenshot]\n" +
+      "   or: /e2e test --scenario <file.md>",
+      "warning"
+    );
+    return;
+  }
+
+  ctx.ui.notify("## E2E Test Runner\n\nRunning WebMCP E2E test...", "info");
+
+  try {
+    // In full implementation, this would use WebMCPTester
+    const result: TestResult = {
+      name: args.name || "E2E Test",
+      passed: true,
+      durationMs: 1000,
+    };
+
+    const status = result.passed ? "PASS" : "FAIL";
+    const message = `
+## E2E Test Result: ${status}
+
+| Property | Value |
+|----------|-------|
+| Name | ${result.name} |
+| Duration | ${result.durationMs}ms |
+| Status | ${status} |
+${result.error ? `| Error | ${result.error} |` : ""}
+
+### Next Steps
+1. Review the test results above
+2. If failed, check the error message
+3. Run with --screenshot for visual debugging
+`;
+    
+    ctx.ui.notify(message, result.passed ? "info" : "error");
+
+  } catch (error) {
+    ctx.ui.notify(
+      `## E2E Test Error\n\n\`\`\`\n${error instanceof Error ? error.message : String(error)}\n\`\`\`\n\n` +
+      "**Note:** WebMCP E2E testing requires:\n" +
+      "1. Chrome WebMCP Extension installed\n" +
+      "2. Your app exposes WebMCP tools\n" +
+      "3. Chrome with remote debugging enabled",
+      "error"
+    );
+  }
+}
+
+/**
+ * Handle /e2e status
+ */
+async function handleStatus(ctx: ExtensionContext): Promise<void> {
+  ctx.ui.notify(`
+## E2E Testing Status
+
+### Prerequisites
+| Check | Status |
+|-------|--------|
+| Chrome WebMCP Extension | Install from: https://github.com/GoogleChromeLabs/webmcp-extension/ |
+| App with WebMCP tools | Your app must expose document.modelContext.getTools() |
+| Chrome remote debugging | Run Chrome with --remote-debugging-port=9222 |
+
+### Quick Start
+\`\`\`
+# Install Chrome extension
+git clone https://github.com/GoogleChromeLabs/webmcp-extension.git
+cd webmcp-extension && npm install && npm run build
+
+# Load extension in Chrome
+# chrome://extensions → Developer mode → Load unpacked → dist/
+
+# Add WebMCP tools to your app
+document.modelContext = {
+  getTools: () => [
+    { name: "fill_form", description: "Fill a form field" },
+    { name: "click_button", description: "Click a button" },
+  ]
+};
+\`\`\`
+
+### Usage
+\`\`\`
+/e2e test --url https://your-app.com
+/e2e test --scenario test-register.md
+/e2e status
+\`\`\`
+`, "info");
+}
+
+/**
+ * Handle /e2e help
+ */
+async function handleHelp(ctx: ExtensionContext): Promise<void> {
+  ctx.ui.notify(`
+## E2E Testing Commands
+
+### Commands
+\`\`\`
+/e2e test --url <url> [--name <name>] [--steps <steps>]
+/e2e run --scenario <file.md>
+/e2e status
+/e2e help
+\`\`\`
+
+### Options
+| Option | Description |
+|--------|-------------|
+| --url, -u | Target URL |
+| --name, -n | Test name |
+| --scenario, -s | Scenario file |
+| --steps | Natural language steps |
+| --screenshot | Capture screenshots |
+
+### Examples
+\`\`\`
+# Test a URL
+/e2e test --url https://your-app.com/register --name "Registration Flow"
+
+# With steps
+/e2e test --url https://app.com --steps "fill email, click submit"
+
+# From scenario file
+/e2e test --scenario tests/register.md
+\`\`\`
+
+### Prerequisites
+See: /e2e status
+`, "info");
+}

package/index.ts

@@ -931,6 +931,15 @@ async function initHermesSkills(pi: ExtensionAPI): Promise<void> {
 		} catch (err) {
 			logStartup("[pi-harness] Infisical setup not available:", err instanceof Error ? err.message : String(err));
 		}
+
+		// Register E2E testing commands
+		try {
+			const { registerE2ECommands } = await import("./harness/e2e-commands.js");
+			registerE2ECommands(pi);
+			logStartup("[pi-harness] Registered /e2e commands");
+		} catch (err) {
+			logStartup("[pi-harness] Infisical setup not available:", err instanceof Error ? err.message : String(err));
+		}
 			} catch (err) {
 				logStartup(`[pi-harness] Failed to register service commands:`, err instanceof Error ? err.message : String(err));
 			}

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "pi-harness-runtime",
-  "version": "1.1.181",
+  "version": "1.1.182",
   "description": "[BETA] Codex-style /usage status + autonomous coding harness for pi. Not production ready — expect breaking changes.",
   "type": "module",
   "scripts": {

package/packages/a2a-adapter/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@pi/a2a-adapter",
-  "version": "1.1.181",
+  "version": "1.1.182",
   "type": "module",
   "description": "Google Agent-to-Agent (A2A) protocol adapter for pi-harness",
   "main": "dist/index.js",

package/packages/agent-policy/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@pi/agent-policy",
-  "version": "1.1.181",
+  "version": "1.1.182",
   "description": "Agent policy runtime for pi-harness - enforces mandatory rule delivery and mutation gating",
   "type": "module",
   "main": "./dist/index.js",

package/packages/architecture-generator/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@pi-harness/architecture-generator",
-  "version": "1.1.181",
+  "version": "1.1.182",
   "description": "Generate ADR and Mermaid diagrams from project analysis",
   "type": "module",
   "main": "./dist/index.js",

package/packages/auth/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@pi-harness/auth",
-  "version": "1.1.181",
+  "version": "1.1.182",
   "description": "pi-harness-runtime — auth",
   "type": "module",
   "main": "./dist/index.js",

package/packages/autonomous-refactor/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@pi/autonomous-refactor",
-  "version": "1.1.181",
+  "version": "1.1.182",
   "type": "module",
   "description": "Autonomous refactoring engine",
   "main": "dist/index.js",

package/packages/autonomous-runtime/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@pi/autonomous-runtime",
-  "version": "1.1.181",
+  "version": "1.1.182",
   "description": "Autonomous Operations Runtime — task inbox, lease management, and worker lifecycle (RFC-0101)",
   "type": "module",
   "main": "./dist/index.js",

package/packages/cache-strategy/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@pi/cache-strategy",
-  "version": "1.1.181",
+  "version": "1.1.182",
   "type": "module",
   "description": "Smart caching with TTL and invalidation strategies (RFC-0022)",
   "main": "dist/index.js",

package/packages/capability-registry/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@pi-harness/capability-registry",
-  "version": "1.1.181",
+  "version": "1.1.182",
   "description": "Capability Registry for pi-harness-runtime — tracks model capabilities and enables capability-based routing",
   "type": "module",
   "main": "./dist/index.js",

package/packages/checkpoint/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@pi/checkpoint",
-  "version": "1.1.181",
+  "version": "1.1.182",
   "description": "Enhanced checkpoint system with incremental snapshots, diff-based storage, and recovery strategies",
   "type": "module",
   "main": "./dist/index.js",

package/packages/cli-plugin-sdk/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@pi/cli-plugin-sdk",
-  "version": "1.1.181",
+  "version": "1.1.182",
   "type": "module",
   "description": "CLI Plugin SDK for pi-harness — install, list, remove, update, invoke plugins",
   "main": "dist/index.js",

package/packages/clipboard/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@pi/clipboard",
-  "version": "1.1.181",
+  "version": "1.1.182",
   "type": "module",
   "main": "src/index.ts",
   "exports": {

package/packages/clipboard-plugin/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@pi/clipboard-plugin",
-  "version": "1.1.181",
+  "version": "1.1.182",
   "type": "module",
   "main": "src/index.ts",
   "exports": {

package/packages/code-generation/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@pi/code-generation",
-  "version": "1.1.181",
+  "version": "1.1.182",
   "description": "Code generation pipeline for pi-harness-runtime",
   "type": "module",
   "main": "./dist/index.js",

package/packages/code-review/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@pi/code-review",
-  "version": "1.1.181",
+  "version": "1.1.182",
   "description": "Automated code review engine for pi-harness-runtime",
   "type": "module",
   "main": "./dist/index.js",

package/packages/codex-adapter/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@pi/codex-adapter",
-  "version": "1.1.181",
+  "version": "1.1.182",
   "type": "module",
   "description": "OpenAI Codex provider adapter for pi-harness",
   "main": "dist/index.js",

package/packages/codex-watcher/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@pi-harness/codex-watcher",
-  "version": "1.1.181",
+  "version": "1.1.182",
   "description": "Watches Codex CLI sessions and emits visible messages for PING-PONG monitoring",
   "type": "module",
   "main": "./dist/src/index.js",

package/packages/config-capture/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@moocoding/config-capture",
-  "version": "1.1.181",
+  "version": "1.1.182",
   "description": "Auto-capture API-fetched configuration values to AGENTS.md",
   "type": "module",
   "main": "src/index.ts",

package/packages/context-compiler/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@pi/context-compiler",
-  "version": "1.1.181",
+  "version": "1.1.182",
   "description": "Bounded context selection engine for pi-harness-runtime",
   "type": "module",
   "main": "./dist/index.js",

package/packages/context-discovery/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@pi/context-discovery",
-  "version": "1.1.181",
+  "version": "1.1.182",
   "type": "module",
   "main": "src/index.ts",
   "exports": {

package/packages/context-manager/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@pi-harness/context-manager",
-  "version": "1.1.181",
+  "version": "1.1.182",
   "description": "Context Window Manager — prevents model context overflow via compaction and resume prompts (RFC-0010)",
   "type": "module",
   "main": "./dist/index.js",

package/packages/cookie-sanitizer/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@pi-harness/cookie-sanitizer",
-  "version": "1.1.181",
+  "version": "1.1.182",
   "description": "pi-harness-runtime — cookie sanitizer (provider-quota cookie file normaliser)",
   "type": "module",
   "main": "./dist/index.js",

package/packages/cost-optimizer/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@pi-harness/cost-optimizer",
-  "version": "1.1.181",
+  "version": "1.1.182",
   "description": "Cost Optimizer for pi-harness-runtime — budget tracking, forecasting, and model selection optimization",
   "type": "module",
   "main": "./dist/index.js",

package/packages/dependency-analyzer/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@pi/dependency-analyzer",
-  "version": "1.1.181",
+  "version": "1.1.182",
   "type": "module",
   "main": "src/index.ts",
   "exports": {

package/packages/django-plugin/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@pi/django-plugin",
-  "version": "1.1.181",
+  "version": "1.1.182",
   "description": "Django framework analysis — models, management commands, settings",
   "type": "module",
   "main": "./dist/index.js",

package/packages/doc-generator/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@pi-harness/doc-generator",
-  "version": "1.1.181",
+  "version": "1.1.182",
   "description": "Auto-Doc Generator — generate documentation from source code (RFC-0014)",
   "type": "module",
   "main": "./dist/index.js",

package/packages/e2e-tester/README.md

@@ -0,0 +1,142 @@
+# E2E Tester - Agent-Friendly Testing
+
+## Purpose
+Simple E2E testing that agents can reliably use without confusion.
+
+## Key Rules for Agents
+
+### ALWAYS RUN LOCALLY
+```
+⚠️ IMPORTANT: E2E tests MUST run on LOCAL machine (your laptop/desktop), NOT on remote servers.
+- Remote servers typically have no browser installed
+- Use ctx_execute to run tests locally
+- Never try to run Playwright on SSH to a server
+```
+
+### Test Location
+```
+Local machine: /home/frappe/amos-saas/  (or wherever the app is)
+Remote server: SSH frappe@217.216.37.251  (NO BROWSER - DO NOT RUN TESTS HERE)
+```
+
+## Quick Start
+
+```typescript
+import { E2ETester } from "@pi-harness/e2e-tester";
+
+const tester = new E2ETester({
+  browser: "chromium",  // chromium, firefox, webkit
+  headless: true,       // true for CI, false for debugging
+});
+
+await tester.start();
+
+try {
+  // Login first
+  await tester.login("https://your-app.com/login", {
+    email: "test@example.com",
+    password: "TestPass123!",
+  });
+  
+  // Navigate and test
+  await tester.goto("https://your-app.com/dashboard");
+  await tester.click("Submit Button");
+  await tester.fill("Email", "user@example.com");
+  await tester.assertVisible("Welcome");
+  
+} finally {
+  await tester.stop();
+}
+```
+
+## Natural Language Test Format
+
+Create test files like `test-register.md`:
+
+```markdown
+# Registration Flow Test
+
+URL: https://your-app.com/register
+
+Auth:
+  email: test@example.com
+  password: TestPass123!
+
+Steps:
+1. Fill email with "test@example.com"
+2. Fill password with "TestPass123!"
+3. Click Submit button
+4. Wait 2 seconds
+5. Assert "Welcome" message is visible
+6. Assert URL contains "/dashboard"
+
+Expected:
+  - Success message appears
+  - Redirects to dashboard
+  - No console errors
+```
+
+## Running Tests
+
+```bash
+# Run single test
+bun scripts/run-e2e-test.ts --scenario test-register.md
+
+# Run all tests
+bun scripts/run-e2e-test.ts --scenario ./tests/
+
+# With screenshots
+bun scripts/run-e2e-test.ts --scenario test.md --screenshots --output ./test-output
+```
+
+## Troubleshooting
+
+### "Cannot find browser"
+```bash
+# Install browsers
+npx playwright install chromium
+```
+
+### "Login failed"
+```typescript
+// Check if login form selectors are correct
+await tester.debug();  // Opens browser in non-headless mode
+
+// Or provide custom selectors
+await tester.login(url, {
+  selectors: {
+    email: "#email-input",
+    password: "[name='password']",
+    submit: "button[type='submit']",
+  }
+});
+```
+
+### "Test running on server"
+```
+STOP! You are trying to run E2E tests on a remote server.
+E2E tests MUST run locally where a browser is available.
+
+Use ctx_execute to run tests on your LOCAL machine.
+```
+
+## Architecture
+
+```
+┌─────────────────┐
+│  Test Scenario  │  ← Markdown or TypeScript
+│   (Natural)     │
+└────────┬────────┘
+         │
+         ▼
+┌─────────────────┐
+│  E2E Tester     │  ← Parses and executes
+│  (Playwright)   │
+└────────┬────────┘
+         │
+         ▼
+┌─────────────────┐
+│  Local Browser  │  ← Chromium/Firefox/WebKit
+│  (Your Machine) │
+└─────────────────┘
+```