$npx -y skills add MagicCube/agentara --skill video-planPlan short-form video, vlog, travel, lifestyle, documentary, or cinematic social video scripts by negotiating duration, choosing a story arc and visual style, then splitting the film into timed scenes of 4-15 seconds. Use this skill whenever the user asks for a video plan, short
| 1 | # Video Plan |
| 2 | |
| 3 | Use this skill to turn a user's short-film idea into a practical, scene-by-scene production plan. The plan should be useful for shooting, editing, AI video generation, or communicating with a photographer/editor. |
| 4 | |
| 5 | The final plan should be written in the user's preferred local language. If the user writes in Chinese, output the plan in Chinese unless they explicitly ask for another language. Keep the skill instructions in English, but adapt the deliverable language to the user. |
| 6 | |
| 7 | ## Primary Goals |
| 8 | |
| 9 | Create a complete short-video plan that: |
| 10 | |
| 11 | - Negotiates or infers the total duration. |
| 12 | - Splits the video into scenes, where each scene is at least 4 seconds and no more than 15 seconds. |
| 13 | - Prefer scene durations of 4, 9, 12, or 15 seconds when practical. |
| 14 | - Uses a recognizable story arc, such as 起承转合, setup-conflict-turn-resolution, three-act structure, or hook-build-payoff. |
| 15 | - Recommends several style directions and chooses one based on the user's preference. |
| 16 | - Defines a coherent visual tone using references from photography styles, directors, films, or creator aesthetics. |
| 17 | - Includes shot type, scene description, character action, wardrobe suggestions, scenery/prop details, and editing rhythm for every scene. |
| 18 | |
| 19 | ## First Response Workflow |
| 20 | |
| 21 | When a user asks for a video or short-film plan, first extract any constraints they already gave: |
| 22 | |
| 23 | - Topic or title |
| 24 | - Intended platform and aspect ratio |
| 25 | - Total duration |
| 26 | - Destination or setting |
| 27 | - Characters |
| 28 | - Voiceover or dialogue |
| 29 | - Emotional goal |
| 30 | - Target audience |
| 31 | - Required locations, props, brands, or products |
| 32 | - Preferred language |
| 33 | - Available footage versus plan for future shooting |
| 34 | |
| 35 | If essential details are missing, ask concise questions before writing the final plan. Do not over-interview. Ask only what materially changes the result. |
| 36 | |
| 37 | Ask about total duration unless the user already provides one. Offer practical defaults: |
| 38 | |
| 39 | - 30 seconds: fast, high-density social cut |
| 40 | - 45 seconds: compact narrative with breathing room |
| 41 | - 60 seconds: standard cinematic story arc |
| 42 | - 90 seconds: more reflective, travel-documentary pacing |
| 43 | |
| 44 | If the user does not answer and wants you to proceed, choose 60 seconds by default. |
| 45 | |
| 46 | ## Duration Rules |
| 47 | |
| 48 | Every scene must obey: |
| 49 | |
| 50 | - Minimum scene duration: 4 seconds |
| 51 | - Maximum scene duration: 15 seconds |
| 52 | - Preferred durations: 4, 9, 12, 15 seconds |
| 53 | |
| 54 | After choosing or confirming total duration: |
| 55 | |
| 56 | 1. Estimate the average scene length. |
| 57 | 2. Calculate the approximate scene count. |
| 58 | 3. Adjust scenes so the total duration exactly matches the target duration when possible. |
| 59 | 4. If exact matching is awkward, explain the closest practical total and why. |
| 60 | |
| 61 | Examples: |
| 62 | |
| 63 | - 30 seconds: 5 scenes x 6 seconds, or 4 scenes using 9+9+8+4 |
| 64 | - 45 seconds: 5 scenes x 9 seconds |
| 65 | - 60 seconds: 6-8 scenes, commonly 4+9+12+9+9+12+5 or 4 scenes x 15 seconds |
| 66 | - 90 seconds: 8-10 scenes, often built from 9, 12, and 15 second blocks |
| 67 | |
| 68 | ## Story Arc Planning |
| 69 | |
| 70 | Use a clear narrative arc before writing scenes. Prefer 起承转合 when the user is Chinese-speaking or asks for an emotional/travel story: |
| 71 | |
| 72 | - 起: Hook, premise, emotional question, or first visual image. |
| 73 | - 承: Expand the world, establish routine, relationship, place, or desire. |
| 74 | - 转: Contrast, complication, emotional turn, unexpected detail, fatigue, doubt, memory, or realization. |
| 75 | - 合: Resolution, return, answer, image of completion, or quiet aftertaste. |
| 76 | |
| 77 | For more commercial or platform-native content, use: |
| 78 | |
| 79 | - Hook: First 1-4 seconds must create curiosity. |
| 80 | - Build: Show the experience, transformation, or stakes. |
| 81 | - Turn: Add contrast, obstacle, or specificity. |
| 82 | - Payoff: End with a memorable image, line, or action. |
| 83 | |
| 84 | Always include a brief "Story Arc" section before the scene table. |
| 85 | |
| 86 | ## Style Consultation |
| 87 | |
| 88 | Before the final plan, recommend 3-5 style directions unless the user already has a clear style. Keep them concrete and selectable. |
| 89 | |
| 90 | Useful style directions: |
| 91 | |
| 92 | - Cinematic travel essay: restrained, reflective, natural light, voiceover-led. |
| 93 | - Handheld diary vlog: intimate, imperfect, observational, close to the body. |
| 94 | - Editorial city montage: graphic compositions, faster cuts, architecture, street detail. |
| 95 | - Filmic romance travel: soft light, lingering gestures, warm human details. |
| 96 | - Outdoor documentary: wide landscapes, physical movement, weather, terrain, effort. |
| 97 | - Minimal luxury travel: clean compositions, slow pacing, texture, quiet service details. |
| 98 | - High-energy social cut: fast transitions, rhythmic inserts, bold location changes. |
| 99 | |
| 100 | If the user has not chosen, |