# AgentCut prompts: Design systems

Complete .md files with type, color, easing, durations and safe areas for video work. Put one in your project and every agent session follows the same rules.

Source: https://agentcut.dev/prompts/#design
Written by AgentCut unless credited. Free to copy, edit and use. Run anything that edits a project on a copy first.

---

## 1. Motion graphics brand system

When to use it: Drop it into a Remotion project, a Claude Code folder or an After Effects MCP session so every title, lower third and transition shares the same type, color, easing and timing.

Works with: Remotion Agent Skills (official) (https://agentcut.dev/tools/remotion-agent-skills-official/), After Effects MCP (Engine Room) (https://agentcut.dev/tools/after-effects-mcp-engine-room/), AE MCP Bridge (https://agentcut.dev/tools/ae-mcp-bridge/), HyperFrames (HeyGen) (https://agentcut.dev/tools/hyperframes-heygen/), Motion Graphics Skill Pack (https://agentcut.dev/tools/motion-graphics-skills-charlie947/)

Save as: brand/DESIGN.md

Sources and credit: EBU R 95, Safe areas for 16:9 production (https://tech.ebu.ch/publications/r095): Action safe and graphics safe percentages

~~~~markdown
# DESIGN.md: Motion graphics brand system

Read this before creating any graphic, title, caption or transition. Fill in the brackets once, then never improvise outside these rules.

## 0. Dials
| Dial | Value | Meaning |
|---|---|---|
| Energy | [4] | 1 calm documentary, 5 clean YouTube, 10 hype trailer |
| Density | [3] | 1 one idea on screen, 10 dashboard |
| Motion | [5] | 1 cuts and fades only, 10 everything animates |

## 1. Canvas
- Masters: 1920x1080 (16:9) and 1080x1920 (9:16), [30] fps. Never mix frame rates in one project.
- Graphics safe area: keep text inside the central 90% of width and height (5% margin each side, following EBU R 95 graphics safe). Action safe: 93%.
- 9:16 extra rule: no text in the bottom 20% or the right 12% (app buttons and captions sit there).
- Grid: 12 columns, gutter 24 px at 1920 wide. Left align titles to column 1 by default.

## 2. Color
| Role | Token | Hex |
|---|---|---|
| Background dark | bg-dark | [#111111] |
| Background light | bg-light | [#F5F3EE] |
| Text on dark | ink-on-dark | [#F5F3EE] |
| Text on light | ink-on-light | [#141413] |
| Accent | accent | [#E5484D] |
| Muted | muted | [#8A877F] |
- One accent per frame. Accent marks the single most important word, number or shape.
- Text contrast at least 4.5:1 against what is behind it. Over footage, add a 25% black shadow or a box.
- No neon gradients, no rainbow palettes.

## 3. Type
- Display: [typeface], weight 700, tracking -2%, line height 1.0.
- Text: [typeface], weight 400 to 500, tracking 0, line height 1.25.
- Scale at 1080 tall (multiply by 1.0 for 9:16 width 1080):
| Token | px | Use |
|---|---|---|
| t-hero | 120 | One word or short title |
| t-title | 72 | Section titles |
| t-sub | 44 | Subtitles, lower third name |
| t-body | 32 | Lower third role, labels |
| t-small | 24 | Credits, sources. Never smaller |
- Max 7 words per line on screen. Max 2 lines per graphic.

## 4. Easing
| Token | Use | CSS / Remotion | After Effects |
|---|---|---|---|
| ease-out | Anything entering | cubic-bezier(0.16, 1, 0.3, 1) | outgoing influence 0%, incoming 85% |
| ease-in | Anything leaving | cubic-bezier(0.7, 0, 0.84, 0) | outgoing 85%, incoming 0% |
| ease-in-out | Moves on screen | cubic-bezier(0.65, 0, 0.35, 1) | 60% / 60% |
| spring | Playful pops (Energy 6+) | spring({damping: 200}) for no bounce, damping 12 for a small bounce | Overshoot expression, max 4% |
- Never linear for position or scale. Linear is only for constant drifts and tickers.

## 5. Durations (frames at 30 fps)
| Token | Frames | Use |
|---|---|---|
| d-xs | 6 | Pops, small UI |
| d-s | 10 | Text entrances, wipes |
| d-m | 15 | Panels, lower thirds in |
| d-l | 24 | Big title moves, transitions |
| d-hold-min | 45 | Minimum time any text stays fully visible |
- Exits are about 30% faster than entrances.
- Stagger related items by 3 to 4 frames.
- Reading time: hold text at least 0.3 s per word, never under 1.5 s.

## 6. Components
- Title card: t-hero or t-title, centered or column 1, accent underline optional.
- Lower third: name t-sub, role t-body at 70% opacity, 4 px accent bar, bottom left inside safe area.
- Caption: see the AgentCut caption system.
- Transition: hard cut first. Allowed extras: 10 frame wipe in accent color, 15 frame crossfade for time jumps.

## 7. Never
- Drop shadows heavier than 30% opacity, bevels, lens flares, 3D extrusions.
- More than two typefaces.
- Text touching the frame edge.
- Animating every element at once.
~~~~

---

## 2. Caption and subtitle style system

When to use it: Give it to Claude, an FFmpeg skill or your caption tool so every video gets captions that are readable, consistent and on brand.

Works with: ffmpeg-skill (https://agentcut.dev/tools/ffmpeg-skill/), VectCutAPI (CapCut/Jianying API) (https://agentcut.dev/tools/vectcutapi-capcut-jianying-api/), MCP Bridge for Adobe Premiere Pro (https://agentcut.dev/tools/mcp-bridge-for-adobe-premiere-pro/), Remotion Agent Skills (official) (https://agentcut.dev/tools/remotion-agent-skills-official/)

Save as: brand/CAPTIONS.md

Sources and credit: Netflix Timed Text Style Guide (English USA) (https://partnerhelp.netflixstudios.com/hc/en-us/articles/217350977-English-USA-Timed-Text-Style-Guide): Line length, reading speed and duration limits for the subtitle mode

~~~~markdown
# CAPTIONS.md: Caption and subtitle style system

Two modes. Pick one per video and say which in the job.

## Mode A: Subtitles (accessibility, long form, client work)
- Max 42 characters per line, max 2 lines.
- Reading speed: max 20 characters per second for adult viewers.
- Duration: min 0.83 s (20 frames at 24 fps), max 7 s.
- Gap between captions: at least 2 frames.
- Position: bottom center inside title safe. Move to top only when it would cover burned in text or a face.
- Style: [sans serif], white, 2 px black outline or a 60% black box, no animation.
- Line breaks: after punctuation, before conjunctions and prepositions. Never split a name, a number and its unit, or an article and its noun.
- Include [MUSIC] and speaker labels only when the speaker is off screen or unclear.
- Spelling: use the term list below exactly.

## Mode B: Social captions (Shorts, Reels, TikTok)
- 1 to 4 words at a time, one line, max 28 characters.
- Size: 7 to 10% of frame height on 1080x1920.
- Position: center or just below center. Never in the bottom 20% or the right 12% of the frame.
- Style: heavy weight [typeface], [white] fill, 6 to 10 px dark outline, active word in [accent #FFE600].
- Timing: word appears 1 to 2 frames after it starts in the audio. Never before.
- Motion: entrance pop 90% to 100% scale over 4 frames, or none. Choose one and keep it for the whole video.
- Emphasis: max one highlighted word per caption. Numbers and names first.

## Shared rules
- Never change what the speaker said, except removing filler words (um, uh) in Mode B when asked.
- Keep captions off faces and off on screen text.
- Deliver the caption file ([SRT / VTT / ASS]) next to any burned in version.

## Term list
| Spoken | Write |
|---|---|
| [agent cut] | [AgentCut] |
| [resolve] | [DaVinci Resolve] |
| [add yours] | |
~~~~

---

## 3. YouTube thumbnail system

When to use it: Brief Claude, a designer or an image tool so every thumbnail in a series is readable at phone size and recognizably yours.

Works with: adb-mcp (Photoshop + Premiere) (https://agentcut.dev/tools/adb-mcp-photoshop-premiere/), Remotion Agent Skills (official) (https://agentcut.dev/tools/remotion-agent-skills-official/)

Save as: brand/THUMBNAILS.md

Sources and credit: YouTube Help, Add video thumbnails (https://support.google.com/youtube/answer/72431): Recommended size 1280x720, 16:9

~~~~markdown
# THUMBNAILS.md: YouTube thumbnail system

## Canvas
- 1280x720 px, 16:9 (YouTube's recommended size). Export JPG or PNG.
- Design for the smallest place it shows: check it at 168x94 px. If you cannot read it there, it fails.
- Keep the bottom right 15% free of important content (the video length badge sits there).

## Layout (pick one per video, keep it for the series)
1. Face + 3 words: face on the right third, looking at the text or the viewer, text on the left two thirds.
2. Before / after: split down the middle, left "before", right "after", arrow or line between.
3. Object hero: one big object or screenshot, slightly rotated, text top left.
- One focal point. The eye should land in under a second.

## Text
- Max 4 words. Never repeat the title, add to it.
- Heavy condensed or grotesk typeface: [typeface], all caps or title case, one style for the series.
- Size: cap height at least 12% of the thumbnail height.
- White or [brand color] with a dark stroke or a solid block behind it.

## Color
- Background contrasts with YouTube's white and dark modes: avoid pure white and pure black backgrounds.
- One dominant color plus one accent [#HEX]. Faces stay natural.
- Push saturation and contrast slightly more than in the video, but skin must still look like skin.

## Faces and real images
- Real photos or real screenshots only. Clear emotion that matches the video (curious, surprised, focused). No fake expressions.
- Cut out the subject cleanly and add a 4 to 8 px light outline to separate it from the background.

## Series rules
- Same font, same text position and same accent color across the series so returning viewers recognize it.
- Keep a template file with locked layers for background, subject, text and logo.

## Never
- Misleading images or claims the video does not deliver.
- More than 2 colors in the text.
- Small logos or watermarks that compete with the subject.
~~~~

---

## 4. Social shorts template system (9:16)

When to use it: You make Shorts, Reels or TikToks in series and want Claude Code, Remotion or CapCut to follow one structure for hook, captions, B roll and end card.

Works with: Remotion Agent Skills (official) (https://agentcut.dev/tools/remotion-agent-skills-official/), HyperFrames (HeyGen) (https://agentcut.dev/tools/hyperframes-heygen/), VectCutAPI (CapCut/Jianying API) (https://agentcut.dev/tools/vectcutapi-capcut-jianying-api/), podcli (https://agentcut.dev/tools/podcli/), ffmpeg-skill (https://agentcut.dev/tools/ffmpeg-skill/)

Save as: brand/SHORTS.md

~~~~markdown
# SHORTS.md: Social shorts template system

## Canvas
- 1080x1920, [30] fps, H.264, AAC 48 kHz.
- Length: [20 to 45] s. The first frame must already be interesting.
- Safe zone: keep text and faces out of the top 10%, the bottom 20% and the right 12% (platform UI). Main action lives in the middle 60%.

## Structure
| Part | Time | Rules |
|---|---|---|
| Hook | 0 to 2 s | Strongest line or visual first. Text hook on screen, max 6 words, matches what is said |
| Setup | 2 to 8 s | Why this matters, one sentence |
| Body | 8 s to end minus 4 s | One idea per 3 to 5 seconds. Cut or change framing at least every 3 s |
| Payoff | last 4 s | Answer the hook. End on a finished sentence |
| Loop | last frame | Optional: last line flows back into the first |

## Framing
- Speaker: eyes on the upper third line, face never cut by captions.
- Punch ins: 110% to 120% zoom on emphasis, hard cut, no slow zooms on every line.
- B roll and screenshots: full width, top half of the frame, speaker can stay in the bottom half (split layout).

## Captions
- Use CAPTIONS.md Mode B.

## Graphics
- Hook text: [typeface] heavy, 72 to 96 px, [white on accent block].
- Progress bar (optional): 6 px tall at the very top, accent color.
- No end card longer than 1 second. No "like and subscribe" text unless I ask.

## Audio
- Voice around -14 LUFS integrated, music ducked 12 to 18 dB under voice.
- No silence longer than 0.4 s unless it is a deliberate beat.

## Files
- Name: [series]_[number]_[3 word slug].mp4
- Deliver with: the caption file, the hook text, a suggested title and the source timecodes.
~~~~

---

## 5. Titles and lower thirds system for interviews

When to use it: Documentaries, corporate interviews and podcasts with many speakers, where every name, title and location card has to match.

Works with: After Effects MCP (Engine Room) (https://agentcut.dev/tools/after-effects-mcp-engine-room/), AE MCP Bridge (https://agentcut.dev/tools/ae-mcp-bridge/), MCP Bridge for Adobe Premiere Pro (https://agentcut.dev/tools/mcp-bridge-for-adobe-premiere-pro/), Remotion Agent Skills (official) (https://agentcut.dev/tools/remotion-agent-skills-official/), Plainly Videos MCP (official) (https://agentcut.dev/tools/plainly-videos-mcp-official/)

Save as: brand/TITLES.md

~~~~markdown
# TITLES.md: Titles and lower thirds system

## Card types
| Type | Content | Position | Hold |
|---|---|---|---|
| Name strap | Name + role | Bottom left, safe area | 4 to 6 s, first appearance only |
| Location | Place, city | Top left | 3 s |
| Date or time jump | "Three months later" | Center | 2.5 s |
| Chapter | Number + title | Center or left column | 3 s |
| Source | "Source: [name]" | Bottom right, small | As long as the claim is on screen |

## Type
- Name: [typeface] Semibold, 44 px at 1080 tall.
- Role: same typeface Regular, 30 px, 75% opacity.
- Chapter titles: [display typeface or serif], 72 px.
- All caps only for short labels (LOCATION, CHAPTER 2), never for names.

## Layout
- Left edge aligned to 5% of width for every card. Bottom edge at 10% of height for name straps.
- Accent bar: [4] px wide, [accent #HEX], height equals the text block.
- Over busy footage: box at [#111111] 80% opacity, 16 px padding.

## Motion
- In: bar wipes 8 frames, name 12 frames ease out with a 20 px slide, role follows 4 frames later.
- Out: 8 frames ease in, all elements together.
- Same timing for every card type. No exceptions per speaker.

## Content rules
- Show each person's name strap the first time they speak, and again only after a long absence.
- Roles: short and how the person wants to be described. Confirm spelling with the client list.
- Source cards for every number or claim that is not the speaker's own.

## Speaker list
| Name | Role | Pronunciation note |
|---|---|---|
| [Name] | [Role] | |
~~~~

---

## 6. Explainer and data animation system

When to use it: Animated charts and number explainers made with Claude Motion, Remotion or After Effects, where clarity and honest numbers matter more than flash.

Works with: Remotion Agent Skills (official) (https://agentcut.dev/tools/remotion-agent-skills-official/), Remotion MCP App (https://agentcut.dev/tools/remotion-mcp-app/), HyperFrames (HeyGen) (https://agentcut.dev/tools/hyperframes-heygen/), After Effects MCP (Engine Room) (https://agentcut.dev/tools/after-effects-mcp-engine-room/)

Save as: brand/EXPLAINER.md

Sources and credit: Remotion Agent Skills (remotion-dev/skills) (https://github.com/remotion-dev/skills): Includes a charts rule for Remotion; use it for the implementation details

~~~~markdown
# EXPLAINER.md: Explainer and data animation system

## Principles
- One idea per scene. One chart per scene.
- Every number on screen comes from the source file. If a number is missing, show [placeholder], never invent it.
- Show the source on screen, small, bottom right, whenever a number appears.

## Scene timing
- Each scene: 4 to 8 seconds. Build 1 to 1.5 s, hold at least 2 s, exit 0.5 s.
- Numbers count up over 1 s with ease out, then hold the final value.
- Narration leads, animation follows: the visual appears on the word, not before.

## Charts
- Bar charts start at zero. Always.
- Max 6 bars or 4 lines per chart. Group the rest as "Other".
- Highlight one data point in the accent color, everything else in muted gray.
- Labels directly on the bars or line ends, no legends when possible.
- Axis labels and units always visible. Round to what matters (18%, not 18.27%).
- Bars grow from the baseline with a 3 frame stagger, left to right.

## Type and color
- Use the motion brand system (DESIGN.md) for fonts, colors and easing.
- Numbers use tabular figures so they do not jitter while counting.
- Big number scenes: number at t-hero size, label under it at t-body.

## Structure for a 30 to 60 s explainer
1. Hook: the single most surprising number or claim.
2. Context: what we are measuring and since when.
3. Three beats: one chart or diagram each.
4. Takeaway: one sentence, held still for a voiceover.

## Never
- 3D pie charts, dual axes, truncated axes, animated gridlines that distract.
- Stock icons for every word. Use an icon only when it explains faster than text.
~~~~

---

## 7. Color and grade look system

When to use it: A series, channel or client with several editors or an agent touching color, and you want every episode to land in the same look.

Works with: DaVinci Resolve MCP Server (samuelgursky) (https://agentcut.dev/tools/davinci-resolve-mcp-server-samuelgursky/), DaVinci Resolve MCP (440+ tools) (https://agentcut.dev/tools/davinci-resolve-mcp-440-tools/), Resolve Claude MCP (https://agentcut.dev/tools/resolve-claude-mcp/), Wasserman's Unofficial DaVinci MCP (https://agentcut.dev/tools/wasserman-s-unofficial-davinci-mcp/)

Save as: brand/LOOK.md

~~~~markdown
# LOOK.md: Color and grade look system

## Pipeline
- Color managed: DaVinci Wide Gamut / DaVinci Intermediate timeline, output Rec.709 Gamma 2.4 for web and TV. [Change if your deliverable is HDR or P3]
- Input transforms per camera, set once per clip on the INPUT node or via project color management. Never grade log footage by eye without a transform.

## Node template (same order on every clip)
1. INPUT CST (if not project managed)
2. EXPOSURE and WB
3. CONTRAST
4. SAT and HUE
5. SECONDARIES (skin, sky): qualifier windows only here
6. LOOK (shared group node or PowerGrade, never edited per clip)
7. OUTPUT CST

## The look, in words
- Name: [Series look name]
- Contrast: [medium], blacks slightly lifted, no clipped highlights on skin.
- Skin: natural, on the skin tone line of the vectorscope.
- Shadows: [slightly cool]. Highlights: [slightly warm].
- Saturation: [90%] overall, highlights desaturated at the top.
- Texture: [fine grain at low strength] or none.
- References: [3 real frames from your own past episodes, with timecodes].

## Shot matching rules
- Match to the hero shot of each scene, not shot to shot down the timeline.
- Check on scopes: waveform for exposure, vectorscope for skin, parade for balance.
- Faces in a scene stay within a small exposure range of each other.

## For agents
- Only touch nodes 2 to 5 per clip. Never edit node 6 (the LOOK) unless asked.
- Label every node. Save a still before and after each change.
- Studio only tools (some ResolveFX like film grain and halation, and external scripting since 21.1): ask before using.
~~~~

---

## 8. Kinetic typography system

When to use it: Text driven videos (lyric videos, quote videos, promo openers) built in Remotion, HyperFrames or After Effects, where you want rhythm and consistency.

Works with: Remotion Agent Skills (official) (https://agentcut.dev/tools/remotion-agent-skills-official/), HyperFrames (HeyGen) (https://agentcut.dev/tools/hyperframes-heygen/), claude-motion-design (https://agentcut.dev/tools/claude-motion-design-howseen/), After Effects MCP (Engine Room) (https://agentcut.dev/tools/after-effects-mcp-engine-room/)

Save as: brand/KINETIC.md

~~~~markdown
# KINETIC.md: Kinetic typography system

## Rhythm
- Lock to music: [BPM]. At 30 fps, one beat = 1800 / BPM frames (120 BPM = 15 frames).
- Words hit on beats or half beats. Lines change on bars (every 4 beats).
- Hold the most important line for at least 2 bars.

## Type
- One typeface family. Two weights max: [Black] for key words, [Medium] for the rest.
- Size by importance, not by length: key word 2x to 3x the size of connecting words.
- Tracking: -2% to -4% on heavy weights, 0 on medium.
- Never more than 6 words on screen at once.

## Motion vocabulary (pick 3 for the whole piece)
| Move | Frames | Easing |
|---|---|---|
| Snap in | 6 | ease out, overshoot max 3% |
| Mask reveal up | 8 | ease out |
| Scale punch | 8 | 100 to 108 to 100 |
| Slide and stop | 10 | ease out, 40 px |
| Letter stagger | 1 frame per letter | ease out |
| Hard cut | 0 | none |
- Use the same entrance for the same kind of word throughout.

## Color
- Background and text: 2 colors. Accent: 1 color, max one word per line.
- Optional full frame color flash on the drop: accent as background for 2 to 4 frames.

## Layout
- Center stack or left aligned column. Pick one per section.
- Keep everything inside 90% graphics safe. On 9:16, keep text in the middle 60% of height.

## Never
- Spinning text, rainbow gradients, every word animating differently, bouncing on every hit.
~~~~