Add script/TTS step handlers, dashboard, and pipeline UI (#35, #36, #43, #44)

StepHandler abstraction (app/Support/VideoWorkflow/StepHandlers/) plugs
into WorkflowEngine without coupling it to how any given step type is
actually executed:
- ScriptGenerationHandler: builds a prompt from the project's title/
  description (or an explicit outline/tone/keywords/target length in
  step input), generates via an active OpenAI AiProvider, stores the
  result as a 'script' asset.
- TtsVoiceoverHandler: voices step.input.text, or falls back to the
  project's latest script asset, via an active ElevenLabs AiProvider;
  voice_id/model_id are configurable per step. No direct "speed"
  control — ElevenLabs' base TTS endpoint doesn't expose one.

RunWorkflowStepJob resolves a step's handler by type and runs it on
the queue, matching the existing PublishPostJob pattern (including a
stale-dispatch guard: no-op if the step isn't 'in_progress' anymore).

UI, under its own /video-projects routes and nav entry — separate
from /projects per the earlier decision:
- VideoProjects/Index (#43): list projects with status and step
  progress, create new ones.
- VideoProjects/Show (#44): start a project from a WorkflowTemplate or
  a hardcoded default 8-step pipeline; per-step Run/Approve/Reject/
  Retry/Skip actions; auto-queues the next step after any transition
  when a handler exists for its type (steps with no handler yet just
  sit in_progress rather than auto-failing); lists project assets.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

tests/Feature/VideoWorkflow/StepHandlers/ScriptGenerationHandlerTest.php

@@ -0,0 +1,91 @@
+<?php
+
+use App\Models\AiProvider;
+use App\Models\VideoProject;
+use App\Support\AiProviders\AiProviderManager;
+use App\Support\AssetStorage\AssetStorage;
+use App\Support\VideoWorkflow\StepHandlers\ScriptGenerationHandler;
+use App\Support\VideoWorkflow\WorkflowEngine;
+use Illuminate\Support\Facades\Http;
+use Illuminate\Support\Facades\Storage;
+
+beforeEach(function () {
+    Storage::fake();
+    $this->engine = new WorkflowEngine;
+    $this->handler = new ScriptGenerationHandler(app(AiProviderManager::class), new AssetStorage);
+});
+
+test('it generates a script, stores it as an asset, and completes the step', function () {
+    Http::fake([
+        'api.openai.com/*' => Http::response([
+            'choices' => [['message' => ['content' => 'Welcome back to the channel today']]],
+        ]),
+    ]);
+    AiProvider::factory()->create(['provider' => 'openai', 'is_active' => true]);
+
+    $project = VideoProject::factory()->create(['title' => 'Best Coffee Gadgets 2026']);
+    $step = $this->engine->start($project, [['type' => 'script_generation']]);
+
+    $this->handler->handle($step, $this->engine);
+
+    $fresh = $step->fresh();
+    expect($fresh->status)->toBe('completed');
+    expect($fresh->output['word_count'])->toBe(6);
+
+    $asset = \App\Models\Asset::find($fresh->output['asset_id']);
+    expect($asset->type)->toBe('script');
+    expect($asset->video_project_id)->toBe($project->id);
+    expect(Storage::get($asset->disk_path))->toBe('Welcome back to the channel today');
+
+    Http::assertSent(fn ($request) => str_contains($request['messages'][0]['content'], 'Best Coffee Gadgets 2026'));
+});
+
+test('it uses an explicit outline, tone, keywords, and target length when given', function () {
+    Http::fake(['api.openai.com/*' => Http::response(['choices' => [['message' => ['content' => 'script']]]])]);
+    AiProvider::factory()->create(['provider' => 'openai']);
+
+    $project = VideoProject::factory()->create();
+    $step = $this->engine->start($project, [['type' => 'script_generation']]);
+    $step->update(['input' => [
+        'outline' => 'A history of espresso machines',
+        'tone' => 'casual and funny',
+        'seo_keywords' => ['espresso', 'coffee history'],
+        'target_words' => 800,
+    ]]);
+
+    $this->handler->handle($step->fresh(), $this->engine);
+
+    Http::assertSent(function ($request) {
+        $prompt = $request['messages'][0]['content'];
+
+        return str_contains($prompt, 'A history of espresso machines')
+            && str_contains($prompt, 'casual and funny')
+            && str_contains($prompt, 'espresso, coffee history')
+            && str_contains($prompt, 'approximately 800 words');
+    });
+});
+
+test('it fails the step when no active OpenAI provider is configured', function () {
+    $project = VideoProject::factory()->create();
+    $step = $this->engine->start($project, [['type' => 'script_generation']]);
+
+    $this->handler->handle($step, $this->engine);
+
+    expect($step->fresh())
+        ->status->toBe('failed')
+        ->error->toContain('No active OpenAI provider');
+});
+
+test('it fails the step when the AI provider call fails', function () {
+    Http::fake(['api.openai.com/*' => Http::response(['error' => ['message' => 'rate limited']], 429)]);
+    AiProvider::factory()->create(['provider' => 'openai']);
+
+    $project = VideoProject::factory()->create();
+    $step = $this->engine->start($project, [['type' => 'script_generation']]);
+
+    $this->handler->handle($step, $this->engine);
+
+    expect($step->fresh())
+        ->status->toBe('failed')
+        ->error->toBe('rate limited');
+});

tests/Feature/VideoWorkflow/StepHandlers/TtsVoiceoverHandlerTest.php

@@ -0,0 +1,92 @@
+<?php
+
+use App\Models\AiProvider;
+use App\Models\VideoProject;
+use App\Support\AiProviders\AiProviderManager;
+use App\Support\AssetStorage\AssetStorage;
+use App\Support\VideoWorkflow\StepHandlers\TtsVoiceoverHandler;
+use App\Support\VideoWorkflow\WorkflowEngine;
+use Illuminate\Support\Facades\Http;
+use Illuminate\Support\Facades\Storage;
+
+beforeEach(function () {
+    Storage::fake();
+    $this->engine = new WorkflowEngine;
+    $this->assets = new AssetStorage;
+    $this->handler = new TtsVoiceoverHandler(app(AiProviderManager::class), $this->assets);
+});
+
+test('it voices explicit input text and completes the step', function () {
+    Http::fake(['api.elevenlabs.io/*' => Http::response('fake-mp3-bytes', 200, ['Content-Type' => 'audio/mpeg'])]);
+    AiProvider::factory()->create(['provider' => 'elevenlabs', 'config' => ['voice_id' => 'v1']]);
+
+    $project = VideoProject::factory()->create();
+    $step = $this->engine->start($project, [['type' => 'tts_voiceover']]);
+    $step->update(['input' => ['text' => 'Hello and welcome']]);
+
+    $this->handler->handle($step->fresh(), $this->engine);
+
+    $fresh = $step->fresh();
+    expect($fresh->status)->toBe('completed');
+    expect($fresh->output['characters'])->toBe(strlen('Hello and welcome'));
+
+    $asset = \App\Models\Asset::find($fresh->output['asset_id']);
+    expect($asset->type)->toBe('audio');
+    expect($asset->mime_type)->toBe('audio/mpeg');
+    expect(Storage::get($asset->disk_path))->toBe('fake-mp3-bytes');
+
+    Http::assertSent(fn ($request) => $request['text'] === 'Hello and welcome');
+});
+
+test('it falls back to the latest script asset when no input text is given', function () {
+    Http::fake(['api.elevenlabs.io/*' => Http::response('bytes')]);
+    AiProvider::factory()->create(['provider' => 'elevenlabs', 'config' => ['voice_id' => 'v1']]);
+
+    $project = VideoProject::factory()->create();
+    $this->assets->store($project, 'script', 'Old draft', 'txt');
+    $this->assets->store($project, 'script', 'Latest draft', 'txt');
+
+    $step = $this->engine->start($project, [['type' => 'tts_voiceover']]);
+
+    $this->handler->handle($step, $this->engine);
+
+    Http::assertSent(fn ($request) => $request['text'] === 'Latest draft');
+});
+
+test('a voice_id in step input overrides the provider default', function () {
+    Http::fake(['api.elevenlabs.io/*' => Http::response('bytes')]);
+    AiProvider::factory()->create(['provider' => 'elevenlabs', 'config' => ['voice_id' => 'default-voice']]);
+
+    $project = VideoProject::factory()->create();
+    $step = $this->engine->start($project, [['type' => 'tts_voiceover']]);
+    $step->update(['input' => ['text' => 'hi', 'voice_id' => 'custom-voice']]);
+
+    $this->handler->handle($step->fresh(), $this->engine);
+
+    Http::assertSent(fn ($request) => str_contains($request->url(), '/text-to-speech/custom-voice'));
+});
+
+test('it fails the step when there is no text to voice', function () {
+    AiProvider::factory()->create(['provider' => 'elevenlabs', 'config' => ['voice_id' => 'v1']]);
+
+    $project = VideoProject::factory()->create();
+    $step = $this->engine->start($project, [['type' => 'tts_voiceover']]);
+
+    $this->handler->handle($step, $this->engine);
+
+    expect($step->fresh())
+        ->status->toBe('failed')
+        ->error->toContain('No script text available');
+});
+
+test('it fails the step when no active ElevenLabs provider is configured', function () {
+    $project = VideoProject::factory()->create();
+    $step = $this->engine->start($project, [['type' => 'tts_voiceover']]);
+    $step->update(['input' => ['text' => 'hi']]);
+
+    $this->handler->handle($step->fresh(), $this->engine);
+
+    expect($step->fresh())
+        ->status->toBe('failed')
+        ->error->toContain('No active ElevenLabs provider');
+});