Model-specific video workflow for 2.5 in the seedance family. Select exact modes, references, provider constraints and prompt structure before generation, editing or continuation.
Installs into .claude/skills of the current project.
Are you the author of Video Seedance?
Add the live security badge to your README. It updates with every re-scan.
[](https://www.skillsdirectory.com/skills/wanfeng1028-video-seedance)
---
name: video-seedance
description: "Model-specific video workflow for 2.5 in the seedance family. Select exact modes, references, provider constraints and prompt structure before generation, editing or continuation."
compatibility: File-reading agents; actual generation requires the separately verified provider or renderer.
metadata:
version: "1.0.0"
snapshot: "2026-10-01"
---
# Seedance workflow
Read [the playbook](references/guide.md), then [the scoped records](references/records.json). Select the exact backend, version and operation before writing a request.
Record inputs, rights, duration, output constraints and budget. Inspect the live provider schema or renderer dependencies. Treat unknown capabilities as unresolved; never infer them from a related checkpoint or service. Generate only after authorization, preserve job/project provenance, and review actual playback and encoded metadata before delivery.
This is an original routing playbook, not a bundled third-party runtime. Upstream skills and their licenses must be reviewed separately.
