seede-design_skill

This skill helps you generate professional design graphics from text or images using Seede AI, streamlining posters, social media visuals, and UI concepts.
  • Python

2.5k

GitHub Stars

3

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 openclaw/skills --skill seede-design

  • _meta.json469 B
  • README.md2.0 KB
  • SKILL.md4.6 KB

Overview

This skill uses Seede AI to generate professional design graphics from text prompts, reference images, or brand themes. It supports posters, social media graphics, UI mockups, and more, with control over size, model, and uploaded assets. The workflow is asynchronous: create a design task, poll for completion, and retrieve resulting images and edit links.

How this skill works

You create an asynchronous design task via the Seede API including prompt, model, size, and optional asset references. The service processes the task (typically 30–90 seconds); poll the task endpoint for status and then download generated images, project edit links, or HTML outputs. Assets can be uploaded and referenced in prompts, and themes or reference images can be embedded to control style and colors.

When to use it

  • Generate event posters or promotional graphics from brief text prompts.
  • Produce social media images that match a brand or reference photo.
  • Create multiple UI design variations from a single brand theme.
  • Add logos or brand assets to a layout and export high-resolution images.
  • Embed the Seede editor in an app for iterative visual edits.

Best practices

  • Obtain and store the SEEDE_API_TOKEN in an environment variable and send it in the Authorization header.
  • Upload logos and reference images first and reference them in prompts using @SeedeMaterial for consistent placement.
  • Specify exact output dimensions (w/h) to match platform requirements and prefer webp for smaller files.
  • Use @SeedeTheme to lock brand colors and typography hints for consistent visual identity.
  • Poll task status with exponential backoff and 2–3 minute timeout handling; expect 30–90 seconds per task.

Example use cases

  • Create a tech conference poster: prompt a minimalist layout, size 1080x1440, include event logo asset.
  • Generate a set of Instagram carousel cards that follow a brand color theme via @SeedeTheme.
  • Produce alternative UI screens for a mobile app by requesting multiple tasks with variant prompts.
  • Convert a reference moodboard into a social post series using @SeedeReferenceImage to guide style.

FAQ

Design tasks typically complete in 30–90 seconds; implement polling with timeouts and retries.

How do I include my logo or images?

Upload assets via the /asset endpoint, then reference them in prompts using @SeedeMaterial with the uploaded filename or URL.

Which image formats are recommended?

webp is recommended for smaller size and faster loading; outputs may include other formats as provided by the API.

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