jordanurbs

Venice Video Model Routing — AI skill for Claude Code

AI community

Claude Code agent skill: Venice AI image/video generation + intelligent model routing for character consistency across multi-shot content.

How to install Venice Video Model Routing

This entry records only its repository, not the path inside it, so there is no exact command to give. Open jordanurbs/venice-video-model-routing and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Venice Video Model Routing does

Claude Code agent skill: Venice AI image/video generation + intelligent model routing for character consistency across multi-shot content.

Alternatives in AI

Generative Media Skills — Multi-modal Generative Media Skills for AI Agents (Claude Code, Cursor, Gemini CLI) 3k ★
GPT Image2 Ecommerce — GPT-Image-2 驱动的电商素材一键生成 Claude Code Skill E-commerce image generation skill powered by GPT-Image-2 via Codex C 332 ★
Nano Banana 2 Skill — AI image generation CLI powered by Gemini 3 Pro 228 ★

README

Venice Video Model Routing for Character Consistency

A Claude Code agent skill that combines Venice AI media generation tools with intelligent model routing for character consistency. Includes executable Python scripts for image generation, video generation, image editing, and upscaling, plus decision trees for choosing the right model, reference images, frame sources, and prompt format for every shot.

Based on the [venice-ai-media](https://github.com/openclaw/skills/tree/main/skills/nhannah/venice-ai-media) skill by [@nhannah](https://github.com/nhannah) for generation execution, extended with production-tested model routing logic for multi-shot character consistency.

What This Skill Does

**Generation layer** (bundled scripts):

Generate images via nano-banana-pro, flux-2-max, and other Venice image models
Generate videos from images via Kling, Veo, WAN, and Sora models
Edit images with AI (face correction, style changes, inpainting)
Upscale images 1-4x with optional AI enhancement

**Routing layer** (decision trees):

R2V-by-default model routing (characters present = R2V with identity anchoring, no characters = atmosphere)
Attach the right reference images (elements, reference_image_urls, scene_image_urls)
Choose the right starting frame (panel image vs previous video's last frame)
Adapt prompts per model (element tokens, inline descriptions, voice cues)
Refine panels via two-pass pipeline (generate then multi-edit)
Format-specific routing for talk shows, interviews, and panel formats
Duration validation against model specs

Prerequisites

Python 3.10+
Venice API key -- get one free at venice.ai/settings/api

export VENICE_API_KEY="vn_your_key_here"

Quick Start

# Generate an image
python3 scripts/venice-image.py --prompt "a cyberpunk street at night" --model nano-banana-pro

# Generate a video from an image
python3 scripts/venice-video.py --image panel