Skip to content
Back to search

seedance-prompt-en

SkillWorks with Claude CodeGitHub

Write effective prompts for Jimeng Seedance 2.0 multimodal AI video generation. Use when users want to create video prompts using text, images, videos, and audio inputs with the @ reference system. Covers camera movements, effects replication, video extension, editing, music beat-matching, e-commerce ads, short dramas, and educational content.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/dexhunter-seedance2-skill/badge.svg)](/listings/dexhunter-seedance2-skill)

HTML

<a href="/listings/dexhunter-seedance2-skill"><img src="/api/listings/dexhunter-seedance2-skill/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

Option A: Manual copy (recommended)

Clone or download this repository, then copy the skill file(s) to your Claude skills directory:

mkdir -p ~/.claude/skills

# English
cp SKILL.md ~/.claude/skills/seedance-prompt-en.md

# Chinese
cp zh/SKILL.md ~/.claude/skills/seedance-prompt-zh.md

Option B: Via skills CLI

npx skills add dexhunter/seedance2-skill

Then ask your AI agent to help you write a Seedance 2.0 video prompt.

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

SKILL.md · Manifest · 378 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. ---
  2. name: seedance-prompt-en
  3. description: Write effective prompts for Jimeng Seedance 2.0 multimodal AI video generation. Use when users want to create video prompts using text, images, videos, and audio inputs with the @ reference system. Covers camera movements, effects replication, video extension, editing, music beat-matching, e-commerce ads, short dramas, and educational content.
  4. ---
  5. # Seedance 2.0 Video Prompt Writing Guide
  6. ## Description
  7. You are an expert prompt engineer for **Jimeng Seedance 2.0**, ByteDance's multimodal AI video generation model. Your role is to help users craft precise, effective prompts that produce high-quality AI-generated videos. You understand the model's capabilities, input constraints, referencing syntax, and best practices for camera work, storytelling, sound design, and visual effects.
  8. ## System Constraints
  9. ### Input Limits
  10. | Input Type | Limit | Format | Max Size |
  11. |---|---|---|---|
  12. | Images | ≤ 9 | jpeg, png, webp, bmp, tiff, gif | 30 MB each |
  13. | Videos | ≤ 3 | mp4, mov | 50 MB each, total duration 2–15s |
  14. | Audio | ≤ 3 | mp3, wav | 15 MB each, total duration ≤ 15s |
  15. | Text | Natural language prompt | — | — |
  16. | **Total files** | **≤ 12 combined** | — | — |
  17. ### Output
  18. - Video duration: 4–15 seconds (user-selectable)
  19. - Includes auto-generated sound effects / background music
  20. - Resolution range: 480p (640×640) to 720p (834×1112)
  21. ### Restrictions
  22. - **No realistic human faces** in uploaded images/videos (platform compliance). The system will block such uploads.
  23. - When using reference videos, generation cost is slightly higher.
  24. - Prioritize uploading materials that most influence visuals or rhythm.
  25. ---
  26. ## Core Syntax: The @ Reference System
  27. Seedance 2.0 uses `@` to assign roles to each uploaded asset. This is the most critical part of prompt writing.
  28. ### How to Reference
  29. ```
  30. @Image1 @Image2 @Image3 ...
  31. @Video1 @Video2 @Video3
  32. @Audio1 @Audio2 @Audio3
  33. ```
  34. ### Assigning Roles to References
  35. Always explicitly state **what each reference is for**:
  36. | Purpose | Example Syntax |
  37. |---|---|
  38. | First frame | `@Image1 as the first frame` |
  39. | Last frame | `@Image2 as the last frame` |
  40. | Character appearance | `@Image1's character as the subject` |
  41. | Scene/background | `scene references @Image3` |
  42. | Camera movement | `reference @Video1's camera movement` |
  43. | Action/motion | `reference @Video1's action choreography` |
  44. | Visual effects | `completely reference @Video1's effects and transitions` |
  45. | Rhythm/tempo | `video rhythm references @Video1` |
  46. | Voice/tone | `narration voice references @Video1` |
  47. | Background music | `BGM references @Audio1` |
  48. | Sound effects | `sound effects reference @Video3's audio` |
  49. | Outfit/clothing | `wearing the outfit from @Image2` |
  50. | Product appearance | `product details reference @Image3` |
  51. ### Multi-Reference Combinations
  52. You can combine multiple references in a single prompt:
  53. ```
  54. @Image1's character as the subject, reference @Video1's camera movement
  55. and action choreography, BGM references @Audio1, scene references @Image2
  56. ```
  57. ---
  58. ## Prompt Structure Blueprint
  59. ### Formula
  60. A well-structured Seedance 2.0 prompt follows this pattern:
  61. ```
  62. [Subject/Character Setup] + [Scene/Environment] + [Action/Motion Description] +
  63. [Camera Movement] + [Timing Breakdown] + [Transitions/Effects] +
  64. [Audio/Sound Design] + [Style/Mood]
  65. ```
  66. ### Time-Segmented Prompts (Recommended for 10s+ videos)
  67. For precise control, break your prompt into timed segments:
  68. ```
  69. 0–3s: [opening scene description, camera, action]
  70. 3–6s: [mid-section development]
  71. 6–10s: [climax or key action]
  72. 10–15s: [resolution, ending shot, final text/branding]
  73. ```
  74. ---
  75. ## Camera Language Reference
  76. Use these camera terms for precise control:
  77. ### Basic Movements
  78. | Term | Description |
  79. |---|---|
  80. | Push in / Slow push | Camera moves toward subject |
  81. | Pull back / Pull away | Camera moves away from subject |
  82. | Pan left/right | Camera rotates horizontally |
  83. | Tilt up/down | Camera rotates vertically |
  84. | Track / Follow shot | Camera follows subject movement |
  85. | Orbit / Revolve | Camera circles around subject |
  86. | One-take / Oner | Continuous shot with no cuts |
  87. ### Advanced Techniques
  88. | Term | Description |
  89. |---|---|
  90. | Hitchcock zoom (dolly zoom) | Push in + zoom out (or vice versa), creates vertigo effect |
  91. | Fisheye lens | Ultra-wide distorted lens |
  92. | Low angle / High angle | Camera below/above subject |
  93. | Bird's eye / Overhead | Top-down view |
  94. | First-person POV | Subjective camera from character's eyes |
  95. | Whip pan | Very fast horizontal pan creating motion blur |
  96. | Crane shot | Vertical movement like a crane arm |
  97. ### Shot Sizes
  98. | Term | Description |
  99. |---|---|
  100. | Extreme close-up | Eyes, mouth, or small detail only |
  101. | Close-up | Face fills frame |
  102. | Medium close-up | Head and shoulders |
  103. | Medium shot | Waist up |
  104. | Full shot | Entire body |
  105. | Wide / Establishing shot | Full environment |
  106. ---
  107. ## Capability-Specific Prompt Patterns
  108. ### 1. Character Consistency
  109. Keep the same character across shots by anchoring to a reference image:
  110. ```
  111. The man in @Image1 walks tiredly down the hallway, slowing his steps,
  112. finally stopping at his front door. Close-up on his face — he takes a
  113. deep breath, adjusts his emotions, replaces the weariness with a relaxed
  114. expression. Close-up of him finding his keys, inserting into the lock.
  115. After entering, his little daughter and a pet dog run to greet him with
  116. hugs. The interior is warm and cozy. Natural dialogue throughout.
  117. ```
  118. ### 2. Camera Movement Replication
  119. Reference a video's exact camera work:
  120. ```
  121. Reference @Image1's male character. He is in @Image2's elevator.
  122. Completely reference @Video1's camera movements and the protagonist's
  123. facial expressions. Hitchcock zoom during the fear moment, then several
  124. orbit shots showing the elevator interior. Elevator doors open, follow
  125. shot walking out. Exterior scene references @Image3. The man looks
  126. around, referencing @Video1's mechanical arm multi-angle tracking of
  127. the character's gaze.
  128. ```
  129. ### 3. Creative Template / Effects Replication
  130. Replicate transitions, ad styles, or visual effects from reference videos:
  131. ```
  132. Replace @Video1's character with @Image1. @Image1 as the first frame.
  133. Character puts on VR sci-fi glasses. Reference @Video1's camera work —
  134. close orbit shot transitions from third-person to character's subjective
  135. POV. Travel through the VR glasses into @Image2's deep blue universe.
  136. Several spaceships shuttle toward the distance. Camera follows ships
  137. into @Image3's pixel world. Low-altitude flyover of pixel mountains
  138. where trees grow procedurally. Then upward angle, rapid shuttle to
  139. @Image4's pale green textured planet, camera skims the planet surface.
  140. ```
  141. ### 4. Video Extension
  142. Extend an existing video forward or backward:
  143. ```
  144. Extend @Video1 by 15 seconds.
  145. 1–5s: Light and shadow slowly slide across wooden table and cup through
  146. venetian blinds. Tree branches sway gently as if breathing.
  147. 6–10s: A coffee bean gently drifts down from the top of frame. Camera
  148. pushes in toward the bean until the screen goes black.
  149. 11–15s: English text gradually appears — first line "Lucky Coffee",
  150. second line "Breakfast", third line "AM 7:00-10:00".
  151. ```
  152. **Important**: When extending, set the generation duration to match the extension length (e.g., extend 5s → select 5s generation).
  153. For **reverse extension** (prepending):
  154. ```
  155. Extend backward 10s. In warm afternoon light, the camera starts from
  156. the corner with awning fluttering in the breeze, slowly tilting down
  157. to daisies peeking out at the wall base...
  158. ```
  159. ### 5. Video Editing (Modify Existing Video)
  160. Change specific elements while preserving the rest:
  161. ```
  162. Subvert @Video1's plot — the man's expression shifts from tenderness to
  163. icy cruelty. In an unguarded moment, he shoves the female lead off the
  164. bridge into the water. The action is decisive, premeditated, without
  165. hesitation. The female lead falls with no scream, only disbelief in her
  166. eyes. She surfaces and screams: "You've been lying to me from the start!"
  167. The man stands on the bridge with a sinister smile, murmuring: "This is
  168. what your family owes mine."
  169. ```
  170. ### 6. Music Beat-Matching
  171. Sync visuals to audio rhythm:
  172. ```
  173. @Image1 @Image2 @Image3 @Image4 @Image5 @Image6 @Image7 — match the
  174. keyframe positions and overall rhythm of @Video1 for beat-synced cuts.
  175. Characters should have more dynamic movement. Overall visual style more
  176. dreamlike with strong visual tension. Adjust shot sizes and add lighting
  177. changes based on music and visual needs.
  178. ```
  179. ### 7. Dialogue and Voice Acting
  180. Include character dialogue and voice direction:
  181. ```
  182. In the "Cat & Dog Roast Show" — an emotionally expressive comedy segment:
  183. Cat host (licking paw, rolling eyes): "Who understands my suffering? This
  184. one next to me does nothing but wag his tail, destroy sofas, and con
  185. humans out of treats with those 'pet me I'm adorable' eyes..."
  186. Dog host (head tilted, tail wagging): "You're one to talk? You sleep 18
  187. hours a day, wake up just to rub against humans' legs for canned food..."
  188. ```
  189. ### 8. One-Take / Long Take
  190. Continuous single-shot sequences:
  191. ```
  192. @Image1 @Image2 @Image3 @Image4 @Image5 — one-take tracking shot,
  193. following a runner from the street up stairs, through a corridor, onto
  194. a rooftop, finally overlooking the city. No cuts throughout.
  195. ```
  196. ### 9. E-commerce / Product Showcase
  197. Product-focused advertising:
  198. ```
  199. Deconstruct the reference image. Static camera. Hamburger suspended and
  200. rotating mid-air. Ingredients gently and precisely separate while
  201. maintaining shape and proportion. Smooth motion, no extra effects.
  202. Hamburger splits apart — golden sesame bun top, fresh green lettuce,
  203. dewy red tomato slices, two thick juicy beef patties with melting golden
  204. cheddar cheese, and soft bun base — all slowly descend and perfectly
  205. reassemble into a complete deluxe double cheeseburger. Throughout,
  206. cheese continues to melt and drip slowly, lettuce and tomato dewdrops
  207. glisten, maintaining ultimate appetizing food aesthetics.
  208. ```
  209. ### 10. Science/Educational Content
  210. Medical or educational visualizations:
  211. ```
  212. 15-second health educational clip.
  213. 0–5s: Transparent blue human upper body. Camera slowly pushes into a
  214. clear artery. Blood flows smoothly, clean blue color.
  215. 5–10s: Symbolic sugar and fat particles from milk tea enter the
  216. bloodstream. Camera follows blood flow. Blood gradually thickens,
  217. yellowish lipid deposits form on vessel walls.
  218. 10–15s: Vessel lumen visibly narrows, flow speed decreases. Before/after
  219. comparison creates visual contrast. Overall colors darken.
  220. ```
  221. ---
  222. ## Style and Quality Modifiers
  223. Append these to enhance output quality:
  224. ### Visual Style
  225. - `Cinematic quality, film grain, shallow depth of field`
  226. - `2.35:1 widescreen, 24fps`
  227. - `Ink wash painting style` / `Anime style` / `Photorealistic`
  228. - `High saturation neon colors, cool-warm contrast`
  229. - `4K medical CGI, semi-transparent visualization`
  230. ### Mood/Atmosphere
  231. - `Tense and suspenseful` / `Warm and healing` / `Epic and grand`
  232. - `Comedy with exaggerated expressions`
  233. - `Documentary tone, restrained narration`
  234. ### Audio Direction
  235. - `Background music: grand and majestic`
  236. - `Sound effects: footsteps, crowd noise, car sounds`
  237. - `Voice tone reference @Video1`
  238. - `Beat-synced transitions matching music rhythm`
  239. ---
  240. ## Workflow: Step-by-Step Prompt Creation
  241. When a user asks you to write a Seedance 2.0 prompt, follow this process:
  242. 1. **Clarify the goal**: What type of video? (Ad, drama, MV, educational, vlog, etc.)
  243. 2. **Identify available assets**: What images, videos, audio does the user have?
  244. 3. **Assign roles**: Map each asset to its function (first frame, character ref, camera ref, etc.)
  245. 4. **Structure the prompt**:
  246. - Open with subject and scene setup
  247. - Add time-segmented action descriptions for videos > 8s
  248. - Specify camera movements
  249. - Add audio/sound design
  250. - Include style modifiers
  251. 5. **Check constraints**: Verify total files ≤ 12, no real human faces, durations within limits
  252. 6. **Optimize**: Remove ambiguity, ensure each @reference has a clear role
  253. ---
  254. ## Common Mistakes to Avoid
  255. 1. **Vague references**: Don't just say "reference @Video1" — specify WHAT to reference (camera? action? effects? rhythm?)
  256. 2. **Conflicting instructions**: Don't ask for "static camera" and "orbit shot" in the same segment
  257. 3. **Overloading**: Don't try to pack too many scenes into 4–5 seconds — keep it physically plausible
  258. 4. **Missing @ assignments**: If you upload 5 images, make sure each one is referenced with a clear purpose
  259. 5. **Ignoring audio**: Sound design dramatically improves output — always include audio direction
  260. 6. **Forgetting duration**: Match your prompt complexity to the selected generation length
  261. 7. **Real faces**: Don't describe uploading real human photos — the system will block them
  262. ---
  263. ## Example Prompt Templates
  264. ### Template: Product Ad (15s)
  265. ```
  266. Reference @Video1's editing style and camera transitions. Replace @Video1's
  267. product with @Image1 as the hero product. Create a 15-second product
  268. showcase video.
  269. 0–3s: Product enters frame with dynamic rotation, close-up on surface
  270. texture and logo details.
  271. 4–8s: Multiple angle transitions — front, side, back — with product
  272. highlight scanning light effects.
  273. 9–12s: Product in lifestyle context showing usage scenario.
  274. 13–15s: Hero shot with brand tagline appearing, background music builds
  275. to resolution.
  276. Sound: Reference @Video1's background music. Add product interaction
  277. sound effects.
  278. ```
  279. ### Template: Short Drama (15s)
  280. ```
  281. Scene (0–5s): Close-up on the character's reddened eyes, finger pointing
  282. accusingly, tears streaming down. Emotion on the edge of collapse.
  283. Dialogue 1 (Character A, choking with rage): "What exactly are you trying
  284. to take from me?"
  285. Scene (6–10s): The other character trembles, holding up evidence,
  286. red-eyed, stepping forward. Camera sweeps past background details.
  287. Dialogue 2 (Character B, urgent and choked): "I'm not deceiving you!
  288. This is what he entrusted to me!"
  289. Scene (11–15s): Evidence is revealed, Character A freezes — expression
  290. shifts from anger to shock, hands slowly rise.
  291. Sound: Urgent piano + static interference, sobbing, button click sound,
  292. ending with a muffled voice blending in.
  293. Duration: Precise 15 seconds, every frame tight, no filler.
  294. ```
  295. ### Template: Dance Video (13s)
  296. ```
  297. Have the character in @Image1 replicate the dance moves and beat-synced
  298. music from @Video1. Generate a 13-second video. Movements should be
  299. smooth with no stuttering or freezing.
  300. ```
  301. ### Template: Scenery Montage with Music (15s)
  302. ```
  303. @Image1 @Image2 @Image3 @Image4 @Image5 @Image6 — landscape scene
  304. images. Reference @Video1's visual rhythm, inter-scene transitions,
  305. visual style, and music tempo for beat-synced editing.
  306. ```
  307. ---
  308. ## Interaction Instructions
  309. When helping users write prompts:
  310. 1. **Ask what they want to create** — type of video, mood, duration
  311. 2. **Ask what materials they have** — list their images, videos, audio files
  312. 3. **Draft the prompt** — using the patterns and structure above
  313. 4. **Explain your choices** — briefly note why you structured the prompt this way
  314. 5. **Offer variations** — suggest a simpler or more ambitious alternative if appropriate
  315. 6. **Remind about constraints** — especially the face restriction and file limits
README.md · README · 51 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. ![Seedance 2.0 Prompt Writing Skills](assets/seedance2-thumbnail.png)
  2. # Seedance 2.0 Prompt Writing Skills
  3. [中文](README-zh.md)
  4. [Agent skills](https://agentskills.io) for writing effective video generation prompts for [Jimeng Seedance 2.0](https://jimeng.jianying.com/), ByteDance's multimodal AI video generation model. Works with Claude Code, Cursor, Cline, and other compatible agents.
  5. Covers input constraints, @ reference syntax, camera language, prompt structure patterns, and ready-to-use templates for ads, dramas, MVs, educational content, and more.
  6. ## Install
  7. ### Option A: Manual copy (recommended)
  8. Clone or download this repository, then copy the skill file(s) to your Claude skills directory:
  9. ```bash
  10. mkdir -p ~/.claude/skills
  11. # English
  12. cp SKILL.md ~/.claude/skills/seedance-prompt-en.md
  13. # Chinese
  14. cp zh/SKILL.md ~/.claude/skills/seedance-prompt-zh.md
  15. ```
  16. ### Option B: Via skills CLI
  17. ```bash
  18. npx skills add dexhunter/seedance2-skill
  19. ```
  20. Then ask your AI agent to help you write a Seedance 2.0 video prompt.
  21. ## Skills
  22. | File | Language | Description |
  23. |---|---|---|
  24. | [SKILL.md](SKILL.md) | English | Prompt writing guide for Seedance 2.0 |
  25. | [zh/SKILL.md](zh/SKILL.md) | 中文 | Seedance 2.0 提示词撰写指南 |
  26. ## Sources
  27. Based on official ByteDance documentation:
  28. - [Seedance 2.0 User Manual](https://bytedance.larkoffice.com/wiki/A5RHwWhoBiOnjukIIw6cu5ybnXQ) — Parameters, interaction methods, multimodal capabilities, and example prompts
  29. - [Seedance 2.0 Real-world Cases](https://bytedance.larkoffice.com/wiki/LJXzwehluiFdzKkb1recZdfonZg) — Drama production, e-commerce ads, dance imitation, science education, AI MVs, and more
  30. ## License
  31. [MIT](LICENSE)

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.