AV

    AI Visuals by shash.builds

    1

    An art director for AI-made images and video: reviews your result for where it will be used, scores it, and hands back the exact fixed prompt or edit with every change explained.

    Free

    Security scanned

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more
    AV

    AI Visuals by shash.builds

    Example session with this skill installed

    [shares an image] Made this in Midjourney for an Amazon main image. Prompt below. What's wrong?

    • Read your context and instructions
    • Compiled the ai visuals by

    Verdict: 71/100, not ready. Biggest problem: the product has a second, invented zipper.

    What's wrong

    1. Zipper: the bag has two pulls, your product has one. Cause: the prompt never says "single zipper".
    2. Background: soft grey gradient, but the placement needs pure white.

    The fix (complete prompt, paste as is)
    [full corrected prompt and settings]

    What changed in the prompt and why

    • "single center zipper" fixes 1
      ~ "studio gradient" -> "pure white background" fixes 2

    Check in the next result: one zipper pull, white edges with no grey halo.

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

    About this skill

    An art director for AI-made images and video: it reviews your result for where it will be used, scores it, and hands back the exact fixed prompt or edit with every change explained.

    What you get

    1. A verdict: score out of 100, ready or not, and the single biggest problem.
    2. Each problem with where it is, its cause and the cheapest fix that works (a targeted edit, a revised prompt, adding text outside the generator, or switching tools).
    3. The complete corrected prompt, ready to paste, in the style your tool expects.
    4. A script-made list of every word removed, added or moved, each tied to one problem.
    5. What to check in the next result.

    Before and after

    Without the skillWith the skill
    "Looks good, maybe make the lighting warmer"A score out of 100 against the placement, with each problem named, where it is and why it happened
    A rewritten prompt with no explanation of what changedThe complete corrected prompt plus a script-made list of every word removed, added or moved, each tied to one problem
    Hands, text and logos get a quick glanceThey are checked on purpose, since those are the usual AI mistakes
    You guess whether it is readyA fixed ready rule: 85 or above, no criterion below 7, and truthfulness and platform rules at 10

    Examples

    Ask: "Is this AI product photo okay for my Etsy listing?" You get: A scored review for the e-commerce placement, covering whether the product looks like what the buyer will receive, plus the exact fix.

    Ask: "My ChatGPT image keeps adding wrong text to the poster. Fix the prompt." You get: The full corrected prompt, the changed words listed, and a note if exact text is something that tool cannot do reliably, with a different route.

    Ask: "Review this Runway clip for a YouTube Short." You get: A check of the video for motion, consistency and artifacts, with a fix for the prompt.

    Ask: "Generate three versions of this prompt and pick the best." (needs your OpenAI key) You get: Claude asks about the cost first, generates with your key, then reviews and scores each result.

    Tools and placements it knows

    ChatGPT, Midjourney, Gemini, Flux, Ideogram, Firefly, Canva, Sora, Veo, Kling and Runway. Placements: e-commerce, Instagram, LinkedIn, YouTube, other social, ads, websites, decks and print.

    Rules it keeps

    Real products stay real (it generates around a real product photo), no fake people, reviews or proof, and AI labelling where platforms require it.

    Free and MIT licensed. Part of the free shash.builds skill collection by Shashwat Jain.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse All Claude Skills.

    It does not generate images unless you set OPENAI_API_KEY and agree to the cost; otherwise you paste the fixed prompt into your tool and share the new result. It stops after three rounds with no improvement and says what the tool cannot do. It never posts or uploads anything. Platform sizes and rules change, so it says when a current spec should be checked.

    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 buying a skill.

    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 to use

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 13 days ago

    • Free with an Agensi account

    Nothing to install. It runs in your AI agent.

    Optional
    Python 3OpenAI API key

    Reviewing needs nothing. Python and the key are only for generation; set OPENAI_API_KEY in your environment. OpenAI bills that use.

    Permissions

    Allowed Hosts

    api.openai.com (only for optional generation, with your key)

    File Scopes

    Images
    videos
    and
    prompt
    text
    files
    you
    give
    it
    A
    scratch
    folder
    for
    outputs

    Reads the key only from the OPENAI_API_KEY environment variable and never writes it to a file. Every generation costs money on your OpenAI account, so Claude asks first.

    Listed13 days ago

    What's inside

    Frequently Asked Questions

    An art director for AI-made images and video: reviews your result for where it will be used, scores it, and hands back the exact fixed prompt or edit with every change explained.

    This skill uses the open SKILL.md standard. Use it through an MCP-compatible agent connected to Agensi, and check the compatibility information on this page for confirmed agents and required capabilities.

    Connect your AI agent to Agensi through MCP, then ask it to use AI Visuals by shash.builds on your task. Your agent loads the latest approved version each time. Paid skills require a purchase or active subscription. Skill files are not available for download; your licence does not allow you to save, copy or export any part of a skill.

    Free