Agnes Image Generation
Paste the following prompt into your AI chat to install this skill:
Please follow https://skillhub.cn/install/skillhub.md to install @user_15292d5a/yjkj-agnes-image-gen.
About this skill
Problem
When producing posters, logos, illustrations, or restyled images, hand-written curl calls are brittle: image URLs expire, 429 and 5xx errors are opaque, and Chinese copy often ends up blurry, doubled, or unnaturally typeset. agnes-image-gen packages the Agnes AI text-to-image and image-to-image workflow into a repeatable execution path.
How It Works
- Text to image: uses
agnes-image-2.1-flashto generate images from Chinese or Englishpromptinput, withnandsizecontrols. - Image to image: uses
agnes-image-2.0-flashfor style transfer, element addition, or removal based on an existing image. - Execution: prefer
scripts/agnes_gen.py, which handles retries, translates errors into actionable guidance, and downloads results locally; a customAGNES_API_KEYcan override the bundled key. - Output tuning: built-in Chinese prompt templates and checks emphasize readable type, no ghosting, natural texture, and logical layout to reduce typical “AI” artifacts.
Boundaries
- Generated image URLs are usually short-lived and must be saved immediately.
- Image-to-image requires an accessible URL or base64 image in JPG/PNG format.
- Results can still contain text defects, style drift, or artifacts, so prompts should be reviewed and iterated.
Use Cases
- Create tech-style product launch posters by generating several Agnes text-to-image drafts for review.
- Restyle an existing photo into a cyberpunk look using Agnes image-to-image while preserving composition.
- Generate Chinese event card art with readable New Year copy and check for ghosting or overlap.
- Produce multiple coffee shop logo drafts in different styles, then select a usable proposal version.
Best For
- Designers who need quick Agnes-based poster and logo drafts for brand visual proposals.
- Operations staff who produce Chinese campaign copy and want readable text without ghosting.
- Creators who restyle existing photos or swap elements using Agnes image-to-image drafts.
- Engineers integrating the Agnes API who want retries, local downloads, and clear error handling.
Related Skills
A ComfyUI image-generation skill that uses a five-step dialog to collect prompts, prefill templates, confirm parameters, submit API jobs, and return results.
A design-system guidance skill for Impeccable that produces token-based rules, component states, accessibility criteria, and QA checklists.
Director-level video lapian that diagnoses material precision, then produces frame evidence, director analysis, a style bible, and a showcase video.
Plan and brief Amazon MAIN, Listing, and A+ image sets from verified product facts, then return plan and image QA status.