Add Playwright E2E harness and test-only helpers; speed up test conta…

.github/workflows/e2e.yml

@@ -0,0 +1,79 @@
+name: Integration & E2E
+
+on:
+  push:
+    branches: [ main ]
+  pull_request:
+    branches: [ main ]
+
+jobs:
+  test:
+    runs-on: ubuntu-latest
+    timeout-minutes: 45
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Set up Node
+        uses: actions/setup-node@v4
+        with:
+          node-version: 20
+          cache: 'pnpm'
+
+      - name: Install pnpm
+        run: |
+          corepack enable
+          corepack prepare pnpm@9 --activate
+
+      - name: Install frontend deps
+        run: pnpm install --frozen-lockfile
+
+      - name: Set up Docker Buildx
+        uses: docker/setup-buildx-action@v3
+
+      - name: Build integration test image
+        run: |
+          docker build -t catnip:latest -f container/Dockerfile .
+          docker build -t catnip:test -f container/test/Dockerfile.test .
+
+      - name: Start test container
+        working-directory: container/test
+        run: docker-compose -f docker-compose.test.yml up -d --no-build
+
+      - name: Wait for health
+        run: |
+          for i in {1..30}; do
+            if curl -fsS http://localhost:8181/health >/dev/null; then exit 0; fi
+            sleep 2
+          done
+          echo "Server failed to start" >&2
+          docker-compose -f container/test/docker-compose.test.yml logs || true
+          exit 1
+
+      - name: Run API integration tests
+        working-directory: container/test/integration
+        run: |
+          sudo apt-get update && sudo apt-get install -y golang-go
+          export CATNIP_TEST_MODE=1
+          export CATNIP_TEST_SERVER_URL="http://localhost:8181"
+          export CATNIP_TEST_DATA_DIR="$GITHUB_WORKSPACE/container/test/data"
+          go test -v -timeout 30m -tags=integration ./...
+
+      - name: Install Playwright Browsers
+        run: pnpm dlx playwright install --with-deps chromium
+
+      - name: Run Playwright E2E
+        env:
+          CATNIP_TEST_SERVER_URL: http://localhost:8181
+        run: pnpm test:e2e
+
+      - name: Upload Playwright report on failure
+        if: failure()
+        uses: actions/upload-artifact@v4
+        with:
+          name: playwright-report
+          path: playwright-report
+
+      - name: Teardown
+        if: always()
+        working-directory: container/test
+        run: docker-compose -f docker-compose.test.yml down

container/internal/cmd/serve.go

@@ -256,6 +256,15 @@ func startServer(cmd *cobra.Command) {
 	v1.Post("/ports/mappings", portsHandler.SetPortMapping)
 	v1.Delete("/ports/mappings/:port", portsHandler.DeletePortMapping)
 
+	// Test-only helpers (enabled when CATNIP_TEST_MODE=1)
+	if os.Getenv("CATNIP_TEST_MODE") == "1" {
+		testHandler := handlers.NewTestHandler(gitService, claudeMonitor)
+		v1.Post("/test/worktrees/:id/title", testHandler.SimulateTitle)
+		v1.Post("/test/worktrees/:id/file", testHandler.SimulateFile)
+		v1.Post("/test/worktrees/:id/rename", testHandler.SimulateRename)
+		log.Printf("🧪 Test endpoints enabled")
+	}
+
 	// Server info route
 	v1.Get("/info", func(c *fiber.Ctx) error {
 		commit, date, builtBy := GetBuildInfo()

container/internal/handlers/test.go

@@ -0,0 +1,115 @@
+package handlers
+
+import (
+	"io/fs"
+	"os"
+	"path/filepath"
+
+	"github.com/gofiber/fiber/v2"
+	"github.com/vanpelt/catnip/internal/models"
+	"github.com/vanpelt/catnip/internal/services"
+)
+
+// TestHandler exposes test-only helpers when CATNIP_TEST_MODE=1
+type TestHandler struct {
+	gitService    *services.GitService
+	claudeMonitor *services.ClaudeMonitorService
+}
+
+func NewTestHandler(git *services.GitService, monitor *services.ClaudeMonitorService) *TestHandler {
+	return &TestHandler{gitService: git, claudeMonitor: monitor}
+}
+
+type simulateTitleRequest struct {
+	Title string        `json:"title"`
+	Todos []models.Todo `json:"todos"`
+}
+
+// SimulateTitle applies a session title and optional todos to a worktree
+func (h *TestHandler) SimulateTitle(c *fiber.Ctx) error {
+	worktreeID := c.Params("id")
+	if worktreeID == "" {
+		return c.Status(400).JSON(fiber.Map{"error": "worktree id is required"})
+	}
+
+	var req simulateTitleRequest
+	if err := c.BodyParser(&req); err != nil || req.Title == "" {
+		return c.Status(400).JSON(fiber.Map{"error": "title is required"})
+	}
+
+	wt, ok := h.gitService.GetWorktree(worktreeID)
+	if !ok || wt == nil {
+		return c.Status(404).JSON(fiber.Map{"error": "worktree not found"})
+	}
+
+	// Notify title change (updates session service and may trigger rename)
+	h.claudeMonitor.NotifyTitleChange(wt.Path, req.Title)
+
+	// Optionally set todos directly in state for UI
+	if len(req.Todos) > 0 {
+		_ = h.gitService.UpdateWorktreeFields(worktreeID, map[string]interface{}{
+			"todos": req.Todos,
+		})
+	}
+
+	return c.JSON(fiber.Map{"ok": true})
+}
+
+type simulateFileRequest struct {
+	RelPath string `json:"path"`
+	Content string `json:"content"`
+}
+
+// SimulateFile writes a file within the worktree to create a change
+func (h *TestHandler) SimulateFile(c *fiber.Ctx) error {
+	worktreeID := c.Params("id")
+	if worktreeID == "" {
+		return c.Status(400).JSON(fiber.Map{"error": "worktree id is required"})
+	}
+	var req simulateFileRequest
+	if err := c.BodyParser(&req); err != nil || req.RelPath == "" {
+		return c.Status(400).JSON(fiber.Map{"error": "path is required"})
+	}
+	wt, ok := h.gitService.GetWorktree(worktreeID)
+	if !ok || wt == nil {
+		return c.Status(404).JSON(fiber.Map{"error": "worktree not found"})
+	}
+
+	// Ensure directory exists and write file
+	target := filepath.Join(wt.Path, req.RelPath)
+	if err := os.MkdirAll(filepath.Dir(target), fs.FileMode(0755)); err != nil {
+		return c.Status(500).JSON(fiber.Map{"error": err.Error()})
+	}
+	if err := os.WriteFile(target, []byte(req.Content), 0644); err != nil {
+		return c.Status(500).JSON(fiber.Map{"error": err.Error()})
+	}
+
+	// Refresh status so UI picks up dirtiness and diffs
+	_ = h.gitService.RefreshWorktreeStatusByID(worktreeID)
+
+	return c.JSON(fiber.Map{"ok": true})
+}
+
+type simulateRenameRequest struct {
+	Branch string `json:"branch"`
+}
+
+// SimulateRename triggers automatic or custom branch rename
+func (h *TestHandler) SimulateRename(c *fiber.Ctx) error {
+	worktreeID := c.Params("id")
+	if worktreeID == "" {
+		return c.Status(400).JSON(fiber.Map{"error": "worktree id is required"})
+	}
+	var req simulateRenameRequest
+	_ = c.BodyParser(&req)
+
+	wt, ok := h.gitService.GetWorktree(worktreeID)
+	if !ok || wt == nil {
+		return c.Status(404).JSON(fiber.Map{"error": "worktree not found"})
+	}
+
+	if err := h.claudeMonitor.TriggerBranchRename(wt.Path, req.Branch); err != nil {
+		return c.Status(400).JSON(fiber.Map{"error": err.Error()})
+	}
+	return c.JSON(fiber.Map{"ok": true})
+}

container/test/README.md

@@ -142,6 +142,19 @@ The integration tests cover these major areas:
 ./run_integration_tests.sh stop
 ```
 
+### End-to-End (Playwright)
+
+The runner now also executes Playwright UI tests against the same test container:
+
+```bash
+# Ensure container is up, then run UI tests only
+./run_integration_tests.sh start
+pnpm dlx playwright install --with-deps chromium
+CATNIP_TEST_SERVER_URL=http://localhost:8181 pnpm test:e2e
+```
+
+In CI, the `Integration & E2E` workflow builds the test container, runs API integration tests, installs Playwright, and runs E2E against `http://localhost:8181`.
+
 ### Development Workflow
 
 The new architecture provides excellent development experience:

container/test/data/gh_data/auth_status.json

@@ -1,5 +1,5 @@
 {
   "hostname": "github.com",
   "username": "testuser",
-  "scopes": ["repo", "workflow", "user"]
+  "scopes": ["repo", "workflow"]
 }

container/test/docker-compose.test.yml

@@ -16,6 +16,7 @@ services:
     environment:
       - CATNIP_TEST_MODE=1
       - PORT=8181
+      - CATNIP_TEST_AIR=${CATNIP_TEST_AIR:-0}
       # Speed up cache operations for faster tests
       - CATNIP_CACHE_DEBOUNCE_MS=50
       - CATNIP_CACHE_BATCH_MS=25

container/test/mocks/claude

@@ -42,6 +42,38 @@ send_title_escape() {
     echo "🪧 New terminal title detected: $title" >> "$MOCK_LOG"
 }
 
+# Write a minimal TodoWrite JSONL entry under ~/.claude/projects for current workdir
+write_todos_jsonl() {
+    local title="$1"
+    local home_dir="${HOME:-/home/catnip}"
+    local workdir
+    workdir=$(pwd)
+    # Transform workdir to Claude projects dir name: replace / with - and ensure leading dash
+    local transformed
+    transformed=$(echo "$workdir" | sed 's#/#-#g')
+    case "$transformed" in
+        -*) : ;; # already has leading dash
+        *) transformed="-$transformed" ;;
+    esac
+    local project_dir="$home_dir/.claude/projects/$transformed"
+    mkdir -p "$project_dir" 2>/dev/null || true
+
+    # Choose or create a session file
+    local session_file
+    session_file="$project_dir/$(get_session_uuid).jsonl"
+
+    # Current RFC3339 timestamp
+    local now
+    now=$(date -u +"%Y-%m-%dT%H:%M:%SZ")
+
+    # Append a TodoWrite assistant message containing a single todo related to the title
+    cat >>"$session_file" <<EOF
+{"type":"assistant","timestamp":"$now","message":{"role":"assistant","content":[{"type":"tool_use","name":"TodoWrite","input":{"todos":[{"id":"1","content":"$title","status":"pending","priority":"medium"}]}}]}}
+EOF
+
+    echo "📝 Wrote TodoWrite to $session_file for title '$title'" >> "$MOCK_LOG"
+}
+
 # Get mock response for various prompts
 get_response_for_prompt() {
     local prompt="$1"
@@ -167,6 +199,8 @@ handle_test_command() {
             send_title_escape "$new_title"
             echo "🪧 Set terminal title to: $new_title"
             save_session_state "$(get_session_uuid)" "$new_title"
+            # Also write todos to drive UI updates in tests
+            write_todos_jsonl "$new_title"
             ;;
         *)
             return 1  # Not a test command
@@ -310,18 +344,53 @@ main() {
 
     # Determine mode based on parsed flags
     if [[ "$api_mode" == true ]]; then
-        # API mode - subprocess communication
-        echo "API mode detected, streaming mock response" >> "$MOCK_LOG" 2>/dev/null || true
-
-        # Sleep briefly to simulate processing
-        sleep 1
-
-        # Stream the JSON response format expected by subprocess
-        # Use a simple, clean JSON response
-        cat << 'EOF'
-{"type": "assistant", "message": {"role": "assistant", "content": [{"type": "text", "text": "Hello World function created successfully!"}]}}
-EOF
-        echo "Mock sent simple JSON response" >> "$MOCK_LOG" 2>/dev/null || true
+        # API mode - subprocess communication: read one JSON line from stdin
+        echo "API mode detected, reading stdin for prompt" >> "$MOCK_LOG" 2>/dev/null || true
+
+        # Read a single JSON line from stdin
+        local input_line
+        IFS= read -r input_line || true
+        echo "API input: $input_line" >> "$MOCK_LOG" 2>/dev/null || true
+
+        # Extract prompt content from JSON
+        local prompt=""
+        if [[ -n "$input_line" ]]; then
+            prompt=$(echo "$input_line" | python3 -c "
+import json, sys
+try:
+    data = json.load(sys.stdin)
+    if 'message' in data and 'content' in data['message']:
+        print(data['message']['content'])
+except Exception:
+    pass
+" 2>/dev/null)
+        fi
+
+        # Helper to slugify a title into a branch-friendly string
+        slugify_title() {
+            echo "$1" | tr '[:upper:]' '[:lower:]' | sed 's/[^a-z0-9\- ]/ /g' | tr -s ' ' '-' | sed 's/^-\+//; s/\-\+$//' | cut -c1-60
+        }
+
+        # If this is a branch-name generation request, return a valid branch name
+        if [[ "$prompt" == *"Generate a git branch name"* || "$prompt" == *"Based on this coding session title:"* ]]; then
+            local title
+            title=$(echo "$prompt" | sed -n 's/.*Based on this coding session title: \"\(.*\)\".*/\1/p')
+            if [[ -z "$title" ]]; then
+                title="$prompt"
+            fi
+            local branch
+            branch="feature/$(slugify_title "$title")"
+            if [[ -z "$branch" || "$branch" == "feature/" ]]; then
+                branch="feature/mock-branch"
+            fi
+            printf '{"type":"assistant","message":{"role":"assistant","content":[{"type":"text","text":"%s"}]}}\n' "$branch"
+            echo "Mock sent branch name: $branch" >> "$MOCK_LOG" 2>/dev/null || true
+            return
+        fi
+
+        # Default simple JSON response
+        printf '{"type":"assistant","message":{"role":"assistant","content":[{"type":"text","text":"Hello from mock Claude!"}]}}\n'
+        echo "Mock sent generic JSON response" >> "$MOCK_LOG" 2>/dev/null || true
         return
     elif [[ ! -t 0 ]]; then
         # JSON input mode (piped input)