Generate cinematic short-form videos with native lip-synced audio via RunComfy.
npx skills add https://github.com/runcomfy-com/skills --skill seedance-v2
Please help me install this Agent Skill. Skill: seedance-v2 Source: https://github.com/runcomfy-com/skills/tree/main/seedance-v2 Command: npx skills add https://github.com/runcomfy-com/skills --skill seedance-v2
Users need a fast way to generate cinematic short-form videos with native lip-synced audio using Seedance 2.0 Pro, without manually stitching references, timing, and playback.
Tell the skill to create a 9:16 spokesperson-style video using Seedance 2.0 Pro with a prompt describing the shot and dialogue tone, optionally including an image reference for identity.
High-intent search queries and answers about installing and using this skill.
To generate cinematic short videos with native lip-synced audio, use Seedance 2.0 Pro via RunComfy CLI. Submit a JSON schema with your prompt, reference arrays, aspect ratio, and duration to produce synchronized 4–15s video outputs.
You can combine up to 9 images, 3 video clips, and 3 audio references to maintain multimodal continuity. This ensures identity, scene, and voice tone remain coherent across your generated short-form video.
The best way to create a 9:16 spokesperson ad with synchronized dialogue is running the Seedance 2.0 Pro endpoint. Provide a prompt describing the shot and dialogue tone, alongside an optional image reference for identity continuity.
Seedance 2.0 Pro supports multimodal continuity by accepting prompt, image_url, video_url, and audio_url inputs. This allows cohesive short sequences with native lip-synced in-pass audio generation.
Choose Seedance for multimodal cinematic lip-synced shorts requiring native audio. The skill provides routing guidance to switch to HappyHorse, Wan, Kling, or LTX when your specific video generation needs differ from this use case.
Limitations of generating short-form video with Seedance 2.0 Pro include output durations constrained to 4–15 seconds. It also requires RunComfy CLI execution and a specific JSON input schema to properly process references and download produced URLs.
agentspace-soGenerate cinematic lip-synced videos from text and reference media via RunComfy.
prime-skillsGenerate cinematic text-to-video clips with synchronized audio via RunComfy.
anbeimeGenerate 9:16 shopping guide videos with a digital human persona.
doany-aiRoute inputs to avatar-generation models via the RunComfy CLI.
