More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    StyleDNA: Image to Style Skill

    1

    Turn reference images into a reusable style skill with a measured, named palette, an outlier check, precise style traits and a shareable style card.

    $5

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    StyleDNA: Image to Style Skill

    StyleDNA: Image to Style Skill

    Example session with this skill installed

    Make a style skill from the 4 images in ./refs and call it "Mid-Century Flat".

    • Read your context and instructions
    • Compiled the styledna
    • Generated the image

    Summary

    Measured 4 references and built

    Mid-Century Flat, a warm mid-century modern flat illustration style, with a style card.

    Outlier check

    ref-fox.jpg (fit 45) is more detailed, more textured and higher contrast than the others (fine fur versus flat shapes). Keep it or drop it?

    Palette

    ColorHexShare
    warm cream#FDEFC926%
    olive charcoal#54524417%
    deep teal#1E3E3F10%
    mustard yellow#E6A32D10%
    marigold#E79C3E9%
    pale sage#CBCFB47%
    sage green#5B75697%
    coral red#DF532C7%
    seafoam#88AB906%

    Measured character: light, moderate saturation, medium contrast, moderate detail, some grain

    Visual language

    • Visual identity: mid-century modern flat illustration with a warm screen-print feel
    • Lighting: flat, even illumination; form shown with flat value steps, never cast shadows
    • Linework: no outlines; edges are clean boundaries between flat color shapes
    • Texture: subtle paper grain and fine speckle over flat fills
    • Shapes: simplified geometric forms with soft organic accents
    • Rendering: matte flat color fields; stepped bands instead of gradients

    Rules

    Do: build subjects from simple geometric shapes · keep large cream negative space · stepped color bands, no gradients · 4-6 palette colors per image

    • Avoid: black outlines · smooth gradients or glow · photographic detail · glossy or 3D
    • Not part of the style: foxes, pine trees, sailboats, coffee cups, lamp posts

    Files

    mid-century-flat/ with SKILL.md, style.json, style-metrics.json, references/ and style-card.html.

    Next steps

    Install the folder as a skill in your agent, or paste the prompt recipe into any image tool with a new subject. Open style-card.html to share the style with your team.

    ref-fox.jpg

    PNG · 1536×1024

    Generated

    Example file from a real run - the skill writes it into your workspace.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Give your agent a few reference images and get back a reusable style skill: a SKILL.md that your agent (or any image generator) can follow to make new subjects in that look.

    What you get

    • A measured palette (up to 12 colors, near-duplicates merged) with proportions and color names image tools understand
    • An outlier check that flags references that don't fit the set, with reasons ("more detailed than the others")
    • Precise style traits written with a built-in style vocabulary: visual identity, lighting, linework, texture, shapes, composition, background, rendering
    • Generation rules, negative constraints and subject exclusions, so new images borrow the style but not the original subjects
    • A shareable HTML style card with swatches, traits, rules, the prompt recipe and reference thumbnails (prints to PDF)

    Everything runs locally: no uploads, no API keys. Built by the makers of the free StyleDNA desktop app.

    Want prompts for each image tool, a 0–100 style checker and a tune loop? See StyleDNA Pro.

    Prompt guidance, not model training: results depend on the image generator and never exactly reproduce the references. Measurements cover color and image statistics; other traits are the agent's visual reading and should be reviewed. Animated images are not supported. Max 20 references.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 1 day ago

    Requires

    Python

    Permissions

    File Scopes

    Reference
    images
    the
    user
    points
    to
    (read
    only)
    A
    new
    style-slug
    /
    output
    folder

    No network access. Reference images are never uploaded or modified.

    Listed1 day ago

    What's inside

    Frequently Asked Questions

    No. This skill works on its own inside your agent. The free desktop app (github.com/Layfate/StyleDNA) adds a visual editor and a saved style library.

    No. Measurement runs locally and the agent only reads the files you point to.

    No. It records subject exclusions so new images keep the style but not the original characters, objects or logos.

    New in Image & Video