Agent skill

short-prompt-guide

Strategy for creating efficient short-form video prompts. Use when creating filler shots, atmospheric scenes, or quick video clips that don't require full Production Brief methodology. Covers when to go short vs long, format+style upfront rule, and two approaches (Descriptive vs Directive) for compact yet coherent results.

Stars 1
Forks 0

Install this agent skill to your Project

npx add-skill https://github.com/rfxlamia/flow/tree/main/creative-studio/05d-short-prompt-guide

SKILL.md

Short Prompt Guide

Efficient prompting for filler shots and atmospheric scenes.

Core Principle: Format + Style Upfront

The Golden Rule: Always declare format and style FIRST, then describe scene.

Why it works: Establishes aesthetic framework before details. AI interprets description within style context.

Template:

Format & style: [genre/aesthetic]
[Scene description in 1-3 sentences]

Example:

Format & style: Cinematic B-roll, warm color grade
Coffee cup steaming on wooden table at dawn café. 
Slow dolly in, close-up. Warm morning light, soft diffusion. 
Gentle café ambience. (no subtitles)

When to Use Short Prompts

✅ Ideal For:

  • Filler/atmosphere shots
  • Transition moments
  • Quick establishing shots
  • Vibe openers
  • B-roll footage
  • Simple product shots
  • Single-subject scenes

❌ NOT Suitable For:

  • Scenes with dialogue
  • Multiple characters with continuity
  • Complex action sequences
  • Structured narratives
  • Multi-beat choreography

Decision Rule

Use short if: Scene needs <3 sentences to describe
Use long if: Scene needs dialogue, multiple beats, or character continuity

For long prompts, see: long-prompt-guide

Two Approaches

Approach 1: Descriptive Prompt (Painting a Picture)

What it is: Describe scene, character, or situation. AI creatively interprets details and camera work.

Best for: Atmospheric shots, mood pieces, establishing shots

Structure:

Format & style: [aesthetic]
Subject/Action: [1-2 compact sentences]
Setting & time: [when/where]
Camera/Composition: [shot + angle + 1 movement]
Lighting/Mood: [brief]
Audio: [ambience/short dialogue] (no subtitles if needed)

Example:

Format & style: Cinematic filler, anime watercolor aesthetic
Subject/Action: A rapper writes lyrics in small bedroom studio. 
Posters on walls, microphone on desk, headphones around neck.
Setting: Night, warm desk lamp glow
Camera: Medium shot, slow push in
Lighting: Warm practical lights, cool window moonlight
Audio: Soft lo-fi beats, pen scratching paper

Approach 2: Directive Prompt (Giving a Command)

What it is: Direct command with specific actions, cuts, or sequence.

Best for: UGC content, social media clips, product demos, quick reactions

Structure:

Format & style: [aesthetic]
Make a [type] video: [subject] [action sequence]
[Specific cuts/beats]: [timing]
[Audio]: [sounds/dialogue]

Example:

Format & style: UGC reaction video, iPhone 14 Pro Max, vertical 9:16
Make a taste-test reaction: person tries new energy drink
Three fast cuts - open can (crack sound), sip, smile and nod
Audio: Can opening, sip sound, "Wow, that's good!" (no subtitles)

Ultra-Minimal Template

When you need absolute brevity:

[Format+Style]. [Subject] [Action] in [Setting]. [Camera]. [Lighting]. [Audio]. [Constraints]

Example:

Cinematic B-roll. Coffee cup steaming on wooden table at dawn café. 
Slow dolly in, close-up. Warm morning light, soft diffusion. 
Gentle café ambience. (no subtitles)

Note: Even minimal prompts must include the 8 core components from great-prompt-anatomy, just compressed.

When to Add Detail

Stay Minimal When:

  • ✅ Simple subject (single object, person)
  • ✅ Standard action (walking, sitting, pouring)
  • ✅ Common setting (café, street, park)
  • ✅ Familiar style (cinematic, documentary)

Add Detail When:

  • ⚠️ Specific visual needed (exact wardrobe color for continuity)
  • ⚠️ Uncommon action (complex choreography)
  • ⚠️ Unique setting (specific architectural style)
  • ⚠️ Continuity with other shots required

Audio in Short Prompts

Keep It Brief:

  • Ambience: "Soft rainfall, distant traffic"
  • Short dialogue: She says: "Ready?" (no subtitles)
  • Music: "Lo-fi hip hop beats, mellow"
  • Silence: "No background music"

Omit Audio When:

  • Standard ambience implied (café sounds in café setting)
  • Music not critical to scene mood
  • Silence is default assumption

Common Mistakes to Avoid

❌ Too Vague:

"A person doing something interesting"

✅ Specific:

"A chef flambing dessert, blue flame leaping up"


❌ Format Buried:

"Person walks down street. It's a noir style."

✅ Format First:

"Neo-noir style. Person walks down rain-slicked street."


❌ Multiple Movements:

"Dolly in while panning left and tilting up"

✅ One Movement:

"Arc left around subject"

For camera movement vocabulary, see: camera-movements


❌ Missing Constraints:

"Person talking"

✅ With Constraints:

Person says: "Let's go." (no subtitles)

Quick Reference Decision Tree

Need dialogue or character continuity?
├─ YES → Use long-prompt-guide
└─ NO → Continue

Simple filler/atmosphere shot?
├─ YES → Use short prompt
└─ NO → Use long-prompt-guide

Describe scene or give command?
├─ DESCRIBE → Descriptive approach
└─ COMMAND → Directive approach

Know exact format/style?
├─ YES → Start with format+style
└─ NO → Browse examples for inspiration

Examples Library

For 50+ short prompt examples organized by use case, see: references/examples-library.md

Load examples-library.md when:

  • Need inspiration for specific style
  • Learning descriptive vs directive patterns
  • Want to see format variations
  • Exploring different shot types

Stay in SKILL.md when:

  • Understand core principles
  • Just need template reminder
  • Creating familiar shot types

Expand your agent's capabilities with these related and highly-rated skills.

rfxlamia/flow

arch-v

Video production workflow orchestrator for Veo 3. Guides users through creating professional video prompts via two paths - direct text-to-video OR image-to-video pipeline (Imagen 3/4 → Veo 3). Validates prompt completeness, checks conflicts, ensures all mandatory components present. Integrates camera-movements, great-prompt-anatomy, short-prompt-guide, long-prompt-guide, and imagine skills.

1 0
Explore
rfxlamia/flow

camera-movements

Standardized camera movement vocabulary for Veo 3 video generation. Use when creating video prompts that require specific camera movements, cinematography terminology, or when validating camera movement specifications. Provides authoritative reference for 50+ camera movements (Dolly, Arc, Crane, FPV Drone, Whip Pan, etc.) to ensure consistent, production-ready terminology.

1 0
Explore
rfxlamia/flow

storyteller

Transform abstract/metaphorical narrative into concrete visual story structure. USE WHEN: Converting poetic/theatrical narrative from diverse-content-gen into scene-by-scene visual breakdowns ready for screenwriter formatting. PIPELINE POSITION: diverse-content-gen → **storyteller** → screenwriter → production-validator → imagine → arch-v PRIMARY FUNCTION: Bridge the gap between "altar pribadi" (abstract metaphor) and "woman returns daily to same beach spot" (filmable scene). OUTPUT: Scene breakdown with concrete visual actions, preserved emotional core, and story logic documentation.

1 0
Explore
rfxlamia/flow

long-prompt-guide

Production Brief methodology for complex Veo 3 video scenes. Use when creating scenes with dialogue, character continuity, structured settings, or multi-beat sequences. Provides 11-block framework (Format & Tone, Main Subjects, Wardrobe & Props, Location & Framing, Lighting & Palette, Continuity Rules, Actions & Camera Beats, Montage Plan, Dialogue, Sound & Foley, Finish) for professional, replicable results.

1 0
Explore
rfxlamia/flow

great-prompt-anatomy

Essential framework for creating solid Veo 3 prompts. Use when constructing video prompts, validating prompt completeness, or teaching prompt structure. Defines 8 mandatory components (Subject, Setting, Action, Style/Genre, Camera/Composition, Lighting/Mood, Audio, Constraints) that every prompt must include for professional results.

1 0
Explore
rfxlamia/flow

diverse-content-gen

Agent workflow for generating highly diverse creative content using Verbalized Sampling (VS) technique. Use when user requests multiple variations, brainstorming, creative ideas, or when standard prompting produces repetitive outputs. Increases diversity by 1.6-2.1× while maintaining quality. Works for: blog posts, social media captions, stories, campaign ideas, product descriptions, taglines, and open-ended creative tasks.

1 0
Explore

Didn't find tool you were looking for?

Be as detailed as possible for better results