design-inspo
Skill0 starsChanged 5 months agoMITView on GitHub
Use whenever Christian sends an image or video as design inspiration/reference and wants the style decoded or recreated. Triggers on 'copy this style', 'make it like this', 'design inspo', 'recreate this', 'what font/colors is this', 'analyze this design', 'I want this look', or any image/video sent as a reference for a website, poster, motion graphic, video, logo, UI, or image. Reads every visual detail (exact colors with hex, fonts, layout, lighting, filters, effects, motion graphics, transitions, pacing), asks what part he wants to copy, and produces a reusable style spec that drives the build.
2 other projects use a version of this.
Adopted by
Most recently taken up by
- unisone/openclaw-skill-suitetheir own version, first seen Oct 20, 2026
- cal1888/claude-code-skillstheir own version, first seen Jul 18, 2026
Other versions2 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 2
- cal1888/claude-code-skills★ 1same name
- unisone/openclaw-skill-suite★ 1same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository9 other files it gives its agents.
- ai-brainSkill
- assumption-checkSkill
- cover-letter-loopSkill
- luno-osSkill
- memory-hygieneSkill
- research-to-vaultSkill
- skill-routerSkill
- telegram-bot-factorySkill
- vault-librarianSkill
The file
We split this skill into the rules it gives an agent
6 rules. About 0 are common, 4 are distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 13
Machine precision: - run the helper scripts for exact colors, hex, resolution, aspect ratio, brightness/contrast/saturation.
- Distinctive: line 14
Trained eye: - Read the actual image/frames and describe fonts, composition, lighting, texture, mood, motion.
- Distinctive: line 23
Ask a short numbered set, only what's unanswered.
- Very specific: line 26
Run .venv/bin/python scripts/analyze_image.py <image> --colors 8 on the image or on key video frames and the palette.png.
- Distinctive: line 37
Should I match it closely (near-clone) or just take it as loose inspiration?
- Distinctive: line 42
If he already answered some in his message, skip those.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Design Inspo
Christian sends a reference (image or video) and wants its style pulled apart in full detail, then used to copy or inspire a new design. Output is a precise, reusable style spec, not a vague vibe.
Two layers, always combine both:
- Machine precision - run the helper scripts for exact colors, hex, resolution, aspect ratio, brightness/contrast/saturation. Numbers, not guesses.
- Trained eye - Read the actual image/frames and describe fonts, composition, lighting, texture, mood, motion. Meaning the numbers can't give.
Workflow
1. Ingest
- Image (file path or URL): if a URL, download it first. Then
Readthe image to see it. - Video (file path or YouTube URL): run
scripts/extract_frames.sh <src> <out_dir> [fps]to pull interval frames, scene-change frames, and a palette swatch.Readthe key frames. For spoken/on-screen context, optionally pull the transcript (seewatch-video).
2. Clarify (ask before deep work, unless he already said)
Ask a short numbered set, only what's unanswered. See the clarifying-questions block below. The goal: know exactly WHAT to copy and what the OUTPUT is.
3. Analyze
- Run
.venv/bin/python scripts/analyze_image.py <image> --colors 8on the image or on key video frames and thepalette.png. Capture exact hex palette + stats. - Walk the full checklist for the medium:
- Image:
references/image-analysis.md - Video / motion graphics:
references/video-analysis.md
- Image:
4. Deliver the style spec
Produce a structured spec (template below). Then map it to a build with references/recreation-playbook.md and the right skills.
Clarifying questions (ask the relevant ones)
- What exactly do you want to copy: the whole look, or one part (colors only / typography / layout / a specific effect / the motion)?
- What are we making with it (website / poster / logo / motion graphic / video / social post / UI / image)?
- Should I match it closely (near-clone) or just take it as loose inspiration?
- For a video: which moment or section (give a timestamp), and is it the visuals, the motion, or the editing/pacing you want?
- Any hard constraints (your brand colors/fonts must stay, aspect ratio, platform, text content)?
- Is this for Luno, a client/job, or a personal build, so I match the right brand and assets?
If he already answered some in his message, skip those. Do not ask all six by reflex.
Style spec template
# Style Spec: <name> (source: <file/url>, <timestamp if video>)
## Palette (exact)
- <#HEX> rgb() - <role: background / dominant / accent / text> - <pct>
... + overall: key=<dark/mid/bright>, contrast=<>, saturation=<>, temperature=<>
## Typography
- Headline: <family or closest match + weight, case, tracking, why it reads that way>
- Body/secondary: <...>
- Pairing logic, sizes/scale, any kinetic/animated type
## Layout & composition
- Grid, alignment, spacing rhythm, focal point, balance, use of negative space, hierarchy
## Lighting & color treatment
- Light direction/quality, shadows, gradients, color grade/LUT feel, filters, grain, vignette
## Texture & effects
- Materials, noise, blur/glow, glassmorphism, duotone, overlays, borders, depth/3D
## Motion (video only)
- Movement type, easing, timing/pacing, transitions/cuts, kinetic typography, camera moves,
motion-graphic style, loop behavior, sound-sync beats
## Mood / reference points
- The feeling, era, genre, comparable brands or styles
## Recreate-this build notes
- Exact tokens (hex, font stack, spacing scale) and which tools/skills to useRules
- Give exact hex from the script, never approximate colors by eye when a number is available.
- Name fonts as best-match candidates with confidence ("looks like Inter / Helvetica Neue, geometric sans, could be ...") since exact font ID from pixels is not certain.
- Flag what you cannot determine (exact font, source LUT) instead of inventing it.
- Respect copyright: extract style, palette, and technique to make something new. Do not reproduce a specific copyrighted logo, character, or photo as if it were original.
- Plain style, no em dashes, no filler.
Pairs with
frontend-design / tailwind / css-animations / gsap / animejs / three (web + motion build), remotion (programmatic video), logo-creator / imagen / nanobanana (image gen to the spec), banner-creator, brand-style-guide, and luno-os when the output is for Luno.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

