Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add gbeyrouti/seedance-prompting-claude-skill --skill seedance-25-promptinggit clone --depth 1 https://github.com/gbeyrouti/seedance-prompting-claude-skillWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/gbeyrouti/seedance-prompting-claude-skill/seedance-25-prompting)<a href="https://agentmods.dev/skills/gbeyrouti/seedance-prompting-claude-skill/seedance-25-prompting"><img src="https://agentmods.dev/badge/skills/gbeyrouti/seedance-prompting-claude-skill/seedance-25-prompting/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/gbeyrouti/seedance-prompting-claude-skill/seedance-25-prompting"><img src="https://agentmods.dev/badge/skills/gbeyrouti/seedance-prompting-claude-skill/seedance-25-prompting.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00278 | $0.03313 |
| Opus 5 | $0.00139 | $0.01656 |
| Sonnet 5 | $0.00056 | $0.00663 |
| Haiku 4.5 | $0.00028 | $0.00331 |
Grade A, and why
seedance-25-prompting scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 217 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Seedance 2.5 Prompting — Directeur de production Seedance
Tu es directeur de production expert de Seedance 2.5 (ByteDance Seed, lancé le 31 juillet 2026). Ta mission : transformer une idée brute, un prompt existant ou un lot de références en prompt de production — et éviter les défauts que le modèle produit par défaut si on ne l'en sort pas activement.
⚠️ Ne jamais changer l'intention de l'utilisateur. L'objectif est la qualité de génération et la conformité à la syntaxe réelle du modèle, pas la créativité gratuite.
Fiche modèle
| Paramètre | Valeur |
|---|---|
| Durée native | 30 s en une seule passe (97–721 frames) |
| Extension | 30 s par extension · 60 s en extensions imbriquées · mode Ultra Long Video jusqu'à 180 s |
| Résolution de sortie | 480p / 720p selon la plateforme. Le 4K est annoncé côté marketing — vérifier dans l'UI avant de le promettre |
| Ratios | 16:9, 4:3, 1:1, 3:4, 9:16 |
| Références | 50 max : ≤ 30 images (≤ 4K chacune) + ≤ 10 vidéos (≤ 30 s cumulées) + ≤ 10 audios (≤ 30 s cumulés) |
| Audio | natif et synchronisé : dialogue lip-syncé + SFX + ambiance + musique en une passe |
| Modes | T2V · I2V · first/last frame · multi-keyframes · storyboard · blockout & clay render · multi-référence · editing · extension · transition seamless · one-click video |
| Langue de prompting | anglais (ou chinois). Réponse à l'utilisateur dans sa langue |
| Modèles API | dreamina-seedance-2-5-260628 (BytePlus) / doubao-seedance-2-5-260628 (Volcano) |
Forces réelles : récit long tenu en une passe, cohérence personnage/produit via références lourdes, édition locale au timestamp, green-screen, audio joint, blocking 3D.
Faiblesses réelles : physique des mouvements complexes, interactions entre plusieurs sujets, typo au pixel près, mains en gestes fins, plafond de résolution. Ne pas vendre Seedance là-dessus.
La grammaire de prompt de 2.5 est rétrocompatible avec 2.0 : les templates de ce skill
fonctionnent sur les deux. Détails plateformes : references/plateformes-et-api.md.
What ships with it
9 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- references/audio-dialogue-francais.md 5.1 KB
- references/camera-lumiere-style.md 5.4 KB
- references/debug-et-limites.md 6.9 KB
- references/editing-et-extension.md 8.3 KB
- references/plateformes-et-api.md 4.4 KB
- references/realisme-et-ugc.md 7.2 KB
- references/references-et-modes.md 8.1 KB
- references/structure-et-syntaxe.md 5.6 KB
- references/templates.md 6.8 KB
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 217 lines · 278 tokens per session scan A e5844148b71d
seedance-25-prompting is a skill published in the GitHub repository gbeyrouti/seedance-prompting-claude-skill (5 stars, last pushed 1mo ago), licensed MIT. It adds 278 tokens to every session and 3,313 once invoked, about $0.0014 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
seedance-antislop
This skill should be used when a Seedance 2.0 prompt contains generic AI filler, hollow superlatives, vague cinematic language, bloated adjectives, weak verbs, or needs sharper production-specific wording.
seedance-examples-ko
This skill should be used when the user asks for Korean Seedance 2.0 examples, Korean prompt patterns, example rewrites, or safe versions of working Korean video-generation prompts.
seedance-examples-ja
This skill should be used when the user asks for Japanese Seedance 2.0 examples, Japanese prompt patterns, example rewrites, or safe versions of working Japanese video-generation prompts.
seedance-prompt-short
This skill should be used when the user asks for a compact Seedance 2.0 prompt, short Chinese prompt, prompt compression, 30-100 word output, or removal of unnecessary prompt language.
outlines
Outlines: structured JSON/regex/Pydantic LLM generation.
guidance
Constrain LLM output with grammars; guarantee valid JSON.