Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    BrandLoom — Project Branding Studio

    1

    Create a matching project logo visual, cover, and GitHub preview from your repo and brand assets—with approval before anything is generated.

    $5

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    BrandLoom — Project Branding Studio

    BrandLoom — Project Branding Studio

    Example session with this skill installed

    Use $brandloom for my GitHub project: https://github.com/example/my-project.

    Read the README and the two reference images I attached.

    I need

    • a square project logo visual
    • a 2:1 project cover
    • a GitHub Social Preview
    • Chinese and English versions

    Keep my existing company logo unchanged.

    Use the author-anime and TuoTuo characters on the cover, but do not use characters on the square logo visual.

    Ask one question at a time and show me the full plan before generating anything.

    • Read your context and instructions
    • Compiled the brandloom

    I’ve reviewed the README and reference images. Here is the proposed direction.

    Project message

    “Turn scattered project work into one clear system.”

    Planned outputs

    1. Square logo visual
      Size: 1254×1254
      Direction: Clean product-focused scene
      Characters: None
      Main focus: Project mark and short positioning line

    2. Project cover
      Size: 1774×887
      Direction: A working product scene showing the main workflow
      Characters: author-anime and TuoTuo
      Main focus: Project name, key benefit, and product interface

    3. GitHub Social Preview
      Size: 1280×640
      Direction: A simpler version of the approved cover
      Language: English

    Brand rules

    • Keep the supplied company logo unchanged
    • Reuse the approved project mark
    • Add the final text after the background artwork is generated
    • Save every result as a new version
    • Do not generate the cover until the square visual is approved

    Before I generate anything, let’s confirm the main cover message.

    A. Explain what the project is
    B. Focus on the problem it solves
    C. Focus on the result the user gets
    D. Use your own wording

    Recommended: A

    Reply with A, B, C, or D.

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

    About this skill

    Make your project look like one brand A good project can still look unfinished when its logo visual, cover, and GitHub preview do not feel like they belong together. BrandLoom helps you turn your project into one clear visual set without asking you to write the perfect image prompt. It first reads the materials you choose, such as a GitHub repository, README, project notes, screenshots, an existing company logo, a project icon, or character references. It then guides you through the important choices one question at a time. How it works 1. BrandLoom understands what your project does, who it is for, and what message the visuals should communicate. 2. You choose the wording, visual style, typography, company logo, project mark, characters, layout, language, and output size. 3. BrandLoom shows the proposed direction and waits for your approval. 4. It creates the base artwork, adds the approved text and brand assets, checks the result, and saves a new version. Nothing is generated before you approve the direction. What it can create - A 1254×1254 square project logo visual - A 1774×887 horizontal project cover - A 1280×640 GitHub Social Preview - Chinese, English, and bilingual versions - A new project mark or layouts using your existing brand assets - Logo-only, cover-only, edited, localized, and alternate versions - A plan-only result with copy ideas, layout, shot list, and image prompts BrandLoom includes three character choices: author-anime, TuoTuo, and XingBi. They can be used separately or together. You can also bring your own character when you have permission to use it. Why it is useful Image generators often misspell text, change logos, or make every version look unrelated. BrandLoom creates the background artwork first, then adds important text and approved brand assets through a local layout process. Your official company logo is reused from the original file instead of being redrawn. It also keeps previous outputs, so making a new version does not remove the old one. Who it is for BrandLoom is made for: - Independent developers - One-person companies - AI builders - Open-source maintainers - Product designers - Content creators - People launching a GitHub project, Skill, software product, or digital product Before you buy BrandLoom is a guided workflow, not a one-line instant logo generator. It asks you to review the important choices before creating the final visuals. Full image generation works best with Codex CLI and a built-in image generation tool. Local text layout and image composition require Python 3.12 or newer and Pillow 12.3.0. BrandLoom does not create trademark registrations, editable SVG or AI source files, Figma projects, automatic social publishing, or dozens of platform sizes in bulk. Custom logos, characters, and reference images must be yours or used with permission.

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

    BrandLoom works best in Codex CLI and requires Python 3.12 or newer, Pillow 12.3.0, and a built-in image generation tool for final images.

    The default outputs are PNG files in three fixed sizes: 1254×1254, 1774×887, and 1280×640.

    It does not create editable SVG or AI source files, Figma projects, trademark registrations, automatic publishing, or large batches of platform sizes.

    An official company logo must be supplied as an original file and will not be redrawn by the image model.

    Custom characters, logos, and reference images require clear usage permission.

    BrandLoom asks for approval before generation, so it is not designed for unattended one-click output.

    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

    Money-back guarantee

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

    Trust & safety

    Security scan

    Security scan passed 2 months ago

    Accounts
    GitHub
    Software
    Python

    Permissions

    Environment Variables
    Write Files
    Terminal / Shell
    Read Files
    Browser

    Allowed Hosts

    raw.githubusercontent.com

    File Scopes

    .brandloom/**
    README*
    docs/**
    staging/brand-assets/**

    BrandLoom reads only the project information and brand files selected by the user. It runs the included Python and Pillow scripts locally to place text, combine approved logos or characters, check image dimensions, and save new versions. Project state and generated outputs are stored under `.brandloom/`. Approved reusable assets may be stored under `$CODEX_HOME/brandloom/`. Browser access is used only for links supplied by the user, mainly GitHub project pages. The local scripts do not call external APIs, request API keys, upload the user’s asset library, delete original files, or overwrite earlier outputs. The only environment variable read is `CODEX_HOME`, which is used to locate the local BrandLoom data folder.

    Listed2 months ago

    What's inside

    Frequently Asked Questions

    BrandLoom solves the issue of visual inconsistency and misspelled text in AI-generated assets by creating a cohesive set of logos, covers, and GitHub previews that use your actual brand assets and approved text.

    You will need Python 3.12 or newer and the Pillow 12.3.0 library to handle the local text layout and image composition features that ensure your brand assets remain crisp.

    Yes, BrandLoom can either design a new project mark for you or integrate your existing company logo and characters directly into the generated layouts to maintain brand integrity.

    The skill generates a 1254x1254 square logo visual, a 1774x887 horizontal cover, and a 1280x640 GitHub Social Preview, with support for English, Chinese, or bilingual text.

    No, BrandLoom follows a guided workflow where it proposes a visual direction, wording, and layout for your review; it only proceeds to generate the final assets once you have given your approval.

    BrandLoom is designed to work best with Codex CLI and integrated image generation tools to facilitate the transition from brand analysis to final asset creation.

    New in Branding & Design