logo-batch-generator_skill

This skill batch-generates 10 professional logo variants from brand description or image reference, offering 7 styles and saving results to Obsidian.
  • Python

79

GitHub Stars

1

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill zephyrwang6/myskill --skill logo-batch-generator

  • SKILL.md5.8 KB

Overview

This skill batch-generates professional brand logos from either a text brief or a reference image. It produces 10 logo variants per job across seven popular styles, calls an image-generation API, and saves outputs to a defined Obsidian image folder. Triggers include common Chinese phrases mapped to the generation flow.

How this skill works

There are two entry modes: A) text description where the system collects brand name, industry, core idea and recommends styles; B) image reference where the system analyzes shapes, colors and typography then asks the user for intent. After style selection the skill builds ten structured prompts that vary composition, color and iconography, calls the Gemini image model for each prompt, and writes the PNG files into the Obsidian images directory.

When to use it

  • You need a quick set of diverse logo options from a single brief.
  • You have a reference image and want logos that keep or remix its elements.
  • You want consistent output rules (1:1 ratio, 2K, vector-like look) for branding review.
  • You plan iterative refinement: pick favorites then regenerate variations.
  • You want style-driven exploration across seven mainstream logo types.

Best practices

  • Provide a short clear brand brief: name, industry, core values and intended use.
  • Select or confirm 1–2 recommended styles to narrow the generator’s focus.
  • When uploading references, indicate which elements to keep, change, or discard.
  • Use concise English or Chinese brand names; Chinese text rule is enforced when needed.
  • Review generated set, mark favorites, and request targeted iterations instead of full regenerations.

Example use cases

  • Startup needs 10 logo concepts for investor pitch and web header.
  • A cafe uploads an old logo to preserve a mascot while modernizing colors and type.
  • A designer quickly prototypes lettermark and badge options for a client presentation.
  • An app team explores gradient vs flat interpretations for an icon set.
  • A podcast owner requests retro badge and modern line-art variants to compare tones.

FAQ

Files are saved to an Obsidian image folder named with date and short brand ID. Each file follows an indexed name like 01-flat-icon-center.png to keep variants organized.

Can I change resolution, aspect ratio or model?

Defaults are 1:1 ratio and 2K resolution using the configured Gemini model. Those parameters can be adjusted in the generation script before batch runs.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational