Ingredient & Recipe Catalog

(Part 3 of 4 — same chapter in the PDF; split for the web site.)

Usage class recipe-ingredient Summary Text generation using Venice Uncensored model via Morpheus API Gateway for uncensored content generation, including adult content, controversial topics, and other sensitive material HTTP path POST /generations/text/venice-uncensored Text to Video Generation Slug scrypted:recipe:text-to-video Usage class recipe-ingredient Summary Generate video from text prompts using the default video provider with aspect ratio and storage Video Generation - AWS NovaReel Slug scrypted:recipe:video-generation-aws-novareel Usage class recipe-ingredient Summary Generate video using AWS NovaReel with orientation support and store to CDN HTTP path POST /generations/videos/aws-novareel Video Generation - Grok Imagine Image-to-Video Slug scrypted:recipe:video-generation-grok-imagine-i2v Usage class recipe-ingredient Summary Generate videos from images using xAI Grok Imagine Video image-to-video model via FAL.ai with duration, aspect ratio, and resolution support, and store to CDN HTTP path POST /generations/videos/grok-imagine-i2v Video Generation - Hailuo 2.3 Image-to-Video Slug scrypted:recipe:video-generation-hailuo-2-3-i2v Usage class recipe-ingredient 112

Summary Generate videos from images using Hailuo 2.3 image-to-video model via FAL.ai and store to CDN HTTP path POST /generations/videos/hailuo-2-3-i2v Video Generation - Sora 2 Image-to-Video Slug scrypted:recipe:video-generation-sora2-i2v Usage class recipe-ingredient Summary Generate videos from images using OpenAI Sora 2 (non-Pro) via FAL.ai. Lower cost than Pro. Resolution auto/720p, aspect ratio auto/9:16/16:9, duration 4/8/12s. Output stored to CDN. HTTP path POST /generations/videos/sora2-i2v Video Generation - Sora 2 Pro Image-to-Video Slug scrypted:recipe:video-generation-sora2-pro-i2v Usage class recipe-ingredient Summary Generate videos from images using OpenAI’s Sora 2 Pro image-to-video model via FAL.ai with resolution, aspect ratio, and duration support, and store to CDN HTTP path POST /generations/videos/sora2-pro-i2v Video Generation - Google Veo 3.1 Fast Image-to-Video Slug scrypted:recipe:video-generation-veo-3-1-i2v Usage class recipe-ingredient Summary Generate videos from images using Google Veo 3.1 Fast image-to-video model via FAL.ai with aspect ratio, resolution, and audio support, and store to CDN HTTP path POST /generations/videos/veo-3-1-i2v Video Generation - Google Veo 3 Fast Image-to-Video Slug scrypted:recipe:video-generation-veo-3-i2v 113

Usage class recipe-ingredient Summary Generate videos from images using Google Veo 3 Fast image-to-video model via FAL.ai with aspect ratio, resolution, and audio support, and store to CDN HTTP path POST /generations/videos/veo-3-i2v Video Motion Analysis Slug scrypted:recipe:video-motion-analysis Usage class analysis / vision Summary Estimate object motion rate from video using OpenCV optical flow with camera-motion compensation and scene-cut detection HTTP path POST /analysis/video/motion Video Orchestration (Parallel Scenes) Slug scrypted:recipe:video-orchestration Usage class orchestration / media pipeline Summary Multi-scene orchestrated video generation with parallel image creation, stitching, transitions, and audio layering Video Orchestrator v1 Slug scrypted:recipe:video-orchestrator-v1 Usage class orchestration / media pipeline Summary Single-pass intent-to-video pipeline: scene interpretation, per-scene video generation, transition stitching, and fault-tolerant progress tracking Code path (guess) ingredients/video_orchestrator_v1/ Video Stitch Tool Slug scrypted:recipe:video-tool-stitch 114

Usage class orchestration / media pipeline Summary Concatenate video segments with configurable transitions (crossfade, cut) into a single output file via the media stitch engine HTTP path POST /generate/video/tool/stitch Video Tool - Topaz Upscale Slug scrypted:recipe:video-tool-topaz-upscale Usage class recipe-ingredient Summary Upscale video using FAL.ai Topaz (fal-ai/topaz/upscale/video). Float-based upscaling; framerate unchanged by default. Output stored to CDN. HTTP path POST /generations/videos/tools/upscale/topaz AWS Bedrock Model Registry Slug scrypted:registry:aws-bedrock-models Usage class AWS / Bedrock Summary Authoritative registry of AWS Bedrock model specifications including context window sizes, pricing, and capabilities. Provides accurate model metadata for cost calculation, token management, and model selection. Permissive Prompt Rewriter Slug scrypted:rewriter:permissive-prompt Usage class other / core Summary Permissive Prompt Rewriter ingredient handler - intelligently sanitizes flagged prompts. Content Moderation Router Slug scrypted:router:content-moderation Usage class other / core 115

Summary The Content Moderation Router is a decision-making ingredient that interprets AWS Bedrock Guardrails responses and translates them into standardized routing decisions for downstream workflow processing. AWS Bedrock Guardrails Slug scrypted:text:guardrails Usage class moderation / safety Summary The AWS Bedrock Guardrails ingredient provides content moderation and safety filtering capabilities using Amazon Bedrock Guardrails. This ingredient can analyze both text and image content to detect potentially harmful or inappropriate material according to configurable safety policies. Text Prefixer Slug scrypted:text:prefixer Usage class other / core Summary Prefixer: Scrypted ingredient package (see code and recipes for interface). Code path (guess) ingredients/prefixer/ Text Suffixer Slug scrypted:text:suffixer Usage class other / core Summary Suffixer: Scrypted ingredient package (see code and recipes for interface). Code path (guess) ingredients/suffixer/ Text Summarizer Slug scrypted:text:summarize Usage class other / core Summary LLM-based text summarization ingredient (legacy registration; implementation may vary by version) 116

Error Utility Slug scrypted:util:error Usage class converter / util Summary The Error Utility ingredient provides a simple way to raise errors with specific codes and messages in recipes. It’s commonly used in conditional steps to fail a recipe execution with a clear error message. Code path (guess) ingredients/util_error/ Token Utilities Slug scrypted:utilities:tokens Usage class converter / util Summary Token estimation and max_tokens calculation for LLM API calls. Provides character-based token approximation with configurable strategies for different use cases. Integrates with Model Registry for accurate context window limits and cost estimation. Wait Slug scrypted:utility:wait Usage class converter / util Summary Wait Ingredient Handler. A simple ingredient that waits for a specified number of milliseconds. Useful for allowing CDN propagation, rate limiting, or other time-based delays. Code path (guess) ingredients/wait/ AWS Nova Reel Video Generator Slug scrypted:video:aws-nova-reel Usage class AWS / Bedrock Summary Provides video generation capabilities using Amazon Nova Reel models through AWS Bedrock for single-shot (6 second) and multi-shot composite (12-120 second) video creation. 117

Video Crop Slug scrypted:video:crop Usage class other / core Summary Crop videos to specified dimensions with configurable anchor positioning and automatic padding with black background when needed. Code path (guess) ingredients/video_crop/ Google Veo 3.1 Fast Image-to-Video Slug scrypted:video:google-veo-3-1-fast-i2v Usage class other / core Summary Generate videos by animating an input image using Google’s Veo 3.1 Fast model. The prompt should describe how to animate the input image, including action, style, camera motion, and ambiance details. Code path (guess) ingredients/google_veo_3_1_fast_i2v/ Google Veo 3.1 Fast Text-to-Video Slug scrypted:video:google-veo-3-1-fast-t2v Usage class other / core Summary Generate videos from text prompts using Google’s Veo 3.1 Fast model. The prompt should be descriptive and clear, including subject, context, action, style, camera motion, composition, and ambiance details. Code path (guess) ingredients/google_veo_3_1_fast_t2v/ Google Veo 3 Fast Image-to-Video Slug scrypted:video:google-veo-3-fast-i2v Usage class other / core Summary Google Veo 3 Fast Image-to-Video ingredient handler. This ingredient provides a provider-agnostic interface for generating videos from images using Google’s Veo 3 Fast model. Currently uses FAL.ai as the default provider. 118

Code path (guess) ingredients/google_veo_3_fast_i2v/ Grok Imagine Image-to-Video Slug scrypted:video:grok-imagine-i2v Usage class other / core Summary Grok Imagine Image-to-Video ingredient handler. This ingredient provides image-to-video generation using xAI’s Grok Imagine Video model via FAL.ai (xai/grok-imagine-video/image-to-video). Code path (guess) ingredients/grok_imagine_i2v/ Kling Video-to-Audio Slug scrypted:video:kling-video-to-audio Usage class other / core Summary Extract and generate audio from videos using Kling’s video-to-audio model. This ingredient takes a video input and generates both a dubbed video with audio and a separate audio file in MP3 format. Perfect for adding sound effects and background music to videos. Code path (guess) ingredients/kling_video_to_audio/ Sora 2 Image-to-Video Slug scrypted:video:sora2-i2v Usage class other / core Summary Sora 2 (plain) Image-to-Video ingredient handler. Uses FAL.ai fal-ai/sora-2/image-to-video (non-Pro). Lower cost than Sora 2 Pro. Supports resolution auto/720p, aspect_ratio auto/9:16/16:9, duration 4/8/12s, and model snapshot selection. Code path (guess) ingredients/sora2_i2v/ Sora 2 Pro Image-to-Video Slug scrypted:video:sora2-pro-i2v 119

Usage class other / core Summary Sora 2 Pro Image-to-Video ingredient handler. This ingredient provides a provider-agnostic interface for generating videos from images using OpenAI’s Sora 2 Pro model. Currently uses FAL.ai as the default provider. Code path (guess) ingredients/sora2_pro_i2v/ Topaz Video Upscale Slug scrypted:video:topaz-upscale Usage class other / core Summary Topaz Video Upscale ingredient handler. Upscales video using FAL.ai fal-ai/topaz/upscale/video. Supports float-based upscaling (e.g. 1.5 for 720p->1080p); framerate unchanged by default. Aws Nova Vision Inference Slug scrypted:vision:aws-nova Usage class analysis / vision Summary Provides vision analysis capabilities using Amazon Nova Pro and Nova Lite models through AWS Bedrock for image content understanding and description. Version Parser (Test) Slug test:version:parser Usage class other / core Summary Test-only ingredient for validating the version parser and registry resolution logic Code path (guess) ingredients/test_version_parser_stub/ Count 118 ingredient interface(s); 84 with ingredients//avb.json overlay. Agent graphs (recipes) Each block lists the recipe slug (metadata id), usageClass, and summary from recipes/*.avb.json when present (else from ingredients//avb.json if the id matches, else JSON metadata / 120

heuristics). When httpPath is set in the overlay, the primary POST route is shown; optional publicBaseUrl yields a documentation URL. Recipes without a dedicated path use the generic recipe execution API with a URL-encoded slug. Source JSON path is given for traceability. Audio Generation - ElevenLabs Dialogue v3 Slug scrypted:recipe:audio-generation-elevenlabs-dialogue-v3 Usage class recipe-ingredient Summary Generate dialogue audio using ElevenLabs Dialogue v3 text-to-dialogue model via FAL.ai and store to CDN HTTP path POST /generations/audio/elevenlabs-dialogue-v3 Source JSON recipes/audio-generation-elevenlabs-dialogue-v3.json Character Extraction Slug scrypted:recipe:character-extraction Usage class recipe-ingredient Summary Extract character reference images (front T-pose, side, back, face) from a scene image using vision analysis and Grok Imagine i2v. Optional free-text to target a specific character (e.g. ’the man with black hair’). HTTP path POST /characters/extraction Source JSON recipes/character-extraction.json Content Rewrite Recipe Slug scrypted:recipe:content-rewrite Usage class recipe-ingredient Summary Rewrites text content to address content policy violations while preserving original intent and structure Source JSON recipes/content_rewrite_plan.json 121

Delula Prompt Analyzer Slug scrypted:delula:prompt-analyzer Usage class other / core Summary Analyze plain intent prompts and extract structured information for recipe configuration, including audio requirements, optimized prompts, and speech data HTTP path POST /api/admin/recipe-creator/analyze Source JSON recipes/delula-prompt-analyzer.json Delula Prompts to Video - Google Veo 3.1 Slug scrypted:recipe:delula-prompts-to-video-google-veo-3-1 Usage class recipe-ingredient Summary Generate videos with optional audio from text prompts using Google Veo 3.1 Fast. Supports portrait, square, and landscape orientations at 1080p resolution. Audio generation is included in the video prompt when audio_prompt is provided. HTTP path POST /generations/videos/delula-veo3-1-fast Source JSON recipes/delula-prompts-to-video-google-veo-3-1.json Delula Prompts to Video - NovaReel Slug scrypted:recipe:delula-prompts-to-video-novareel Usage class recipe-ingredient Summary Generate video from prompts using AWS Canvas for image generation and NovaReel for video generation, with optional orientation cropping and audio enhancement HTTP path POST /generations/videos/delula-novareel Source JSON recipes/delula-prompts-to-video-novareel.json Time Series Forecast Slug scrypted:recipe:forecast 122

Source: transcribed from the compiled Scrypted Network Design whitepaper PDF for web reading. Layout, figures, and pagination may differ from the PDF.