Image-to-SVG Preflight

    1

    A practical SKILL.md workflow for crop coordinates, image-size checks and clean SVG handoffs. Includes concrete resize calculations and an optional local browser editor.

    Free

    Security scanned

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Image-to-SVG Preflight

    Example session with this skill installed

    Use image-to-svg-preflight to plan a logo handoff. The source is a 2400 x 1600 PNG, 1,800,000 bytes. Crop x=400, y=200, width=1200, height=1200; output width 600. The preview is 600 x 400. No image file or image-editing tool is attached: give the exact plan and checks, without claiming you edited or traced an image.

    • Read your context and instructions
    • Compiled the image-to-svg preflight

    Plan: a 600 x 600 transparent raster, followed by SVG tracing if appropriate

    The supplied crop is within the source. Its right edge is 400 + 1200 = 1600, below the source width of 2400; its bottom edge is 200 + 1200 = 1400, below the height of 1600. The crop has a 1:1 aspect ratio, so round(600 x 1200 / 1200) gives an output height of 600.

    ItemSupplied or calculated value
    Original2400 x 1600 PNG; 1,800,000 bytes supplied
    Source pixels3,840,000
    Crop in original pixelsx=400, y=200, w=1200, h=1200
    Target raster600 x 600 PNG; 360,000 pixels
    Preview-to-source scale4 horizontally and vertically
    A 100-pixel horizontal preview drag400 original-image pixels

    Next steps

    1. Keep the original file unchanged. With an authorized local editor, apply the exact crop, then resize it to 600 x 600 and export PNG if transparency is needed.
    2. Inspect the actual output dimensions and the logo's margins. The dimensions alone cannot show whether the crop cuts off the subject or antialiasing.
    3. Trace the prepared raster with your chosen compatible tool. Compare source and trace at 100%, then inspect curves, holes and small islands at 200% or 400% when available.
    4. Confirm the downloaded SVG contains path geometry, then test import into its intended destination application.

    For the optional JPG to SVG browser tool, the supplied source values are within its 20,000,000-byte, 4,000,000-pixel and 8,192-pixel-side limits. Verify the actual file before processing. I calculated this plan from the values provided; no image was attached, so no crop, resize, trace, file export or destination import has been performed.

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

    About this skill

    Prepare a logo or flat illustration before tracing it, rather than fixing a poorly prepared SVG afterward. This free, MIT-licensed SKILL.md gives a compatible agent a clear workflow: preserve the original, calculate crops in source pixels, resize proportionally, inspect traced paths, and report the checks actually performed.

    What you get

    • A readable workflow with explicit crop-boundary formulas and a worked 2400 x 1600 example.
    • A checklist for image dimensions, file size, transparency, edge margins and raster versus vector output.
    • A handoff record that separates planned steps from edits and import checks that actually happened.
    • A useful plan even when the agent has no image-editing tools available.

    Using it

    Ask your agent to prepare a source image for its intended destination. The workflow uses tools already available and authorized in that host; it does not install a converter or add an API. Keep your chosen compatible editor when you already have one.

    For an optional visual handoff, JPG to SVG's browser image editor combines local cropping, resizing, raster export and SVG tracing. The same publisher maintains this Skill and that website. The website is optional, has its own file limits and anonymous usage analytics; the Skill's text does not upload images or grant permission to publish artwork.

    Package and license

    The ZIP contains SKILL.md, README.md and the MIT license. It contains no product core source or customer images. Read the public source and usage notes.

    The example below is an actual calculation and planning response to supplied dimensions. No image file was attached, so it does not claim a generated file or a completed visual inspection.

    Prompt-only workflow; actual edits and visual inspection need existing authorized image tools. The optional linked tracer has separate size limits. No conversion engine or universal client/import guarantee is included.

    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

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 14 days ago

    • Free with an Agensi account

    Permissions

    Read Files
    Write Files

    Read/write access applies only to user-selected source and output files when performing actual edits with authorized host tools. A plan needs no file access. Browser/network use is optional only for a user-chosen interactive editor, not a mandatory account connection. Do not upload private artwork or publish it without permission.

    Listed14 days ago

    What's inside

    Frequently Asked Questions

    It gives the agent a repeatable preparation and handoff checklist: calculate original-pixel crops, preserve aspect ratio, check the selected tracer limits, and record what was actually inspected. Editing and inspection depend on tools already available to that agent; it does not guarantee a trace or prevent every processing error.

    The optional JPG to SVG website currently accepts up to 20,000,000 bytes, 4,000,000 pixels and 8,192 pixels on either side. All three apply there. Other editors and tracers have their own limits; these are not universal SKILL.md restrictions.

    The workflow asks an agent with suitable file-reading tools to verify actual SVG path geometry and distinguish it from an embedded raster. Without a file or inspection tool, it returns a concrete check plan instead of claiming that the check happened.

    Use a host that can load SKILL.md instructions. The text works with your existing authorized image tools or returns a plan without them. JPG to SVG is an optional interactive handoff, not an integrated API or required service. Every desktop client and destination app has not been tested.

    This listing is free. The ZIP contains the workflow in SKILL.md, a README and the MIT license. No conversion engine, product core source, customer artwork or mandatory paid service is included.

    The workflow focuses on preparing and inspecting a handoff. Actual editing requires a suitable authorized host tool; background removal, OCR and node editing are not supplied by this package.

    New in Image & Video