alibaba-image-models
Plan, prompt, call, edit, iterate, and productionize Alibaba Cloud Model Studio image generation with current Wan 2.7 Image, Qwen-Image 2.0, and Z-Image Turbo models. Use for Alibaba/DashScope text-to-image, multi-reference generation, instruction editing, character-consistent image sets, typography and layout work, regional authentication, model selection, API integration, cost/rate-limit planning, troubleshooting, safety, rights, privacy, and output QA.
npx skills add calesthio/generative-media-skills --skill alibaba-image-models --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
What it does
Plan, prompt, call, edit, iterate, and productionize Alibaba Cloud Model Studio image generation using Wan 2.7 Image, Qwen-Image 2.0, and Z-Image Turbo models. The skill covers production workflow, model selection, API integration, cost/rate-limit planning, safety, rights, privacy, and output QA, with emphasis on production contracts, region/key/workspace binding, and gated iteration.
How it works
- Establish the production contract before generation: record deliverable type, required pixel dimensions, text, colors, layout, reference image roles, drift allowances, region/workspace/model, budget, retention, and approval gates.
- Announce provider, model, region, image count, and estimated charge; generate one or two low-cost candidates first; avoid silent changes to model/region/prompt policies.
- Route generation via region- and model-appropriate endpoints and protocols. Use synchronous Wan 2.7 or Qwen-Image 2.0/Z-Image Turbo where applicable; for asynchronous Wan, persist IDs and poll task status until SUCCEEDED.
- Build requests with a single user message; structure with an image array and a single text object; use width*height notation for size.
- Use a gated iteration loop: validate locally, request small samples, select structurally, repair along one axis per pass, finish deterministically, run full QA, and persist results with metadata and deterministic crops when needed.
- Follow detailed prompts for controllable production, including production heuristics, layout, typography, and branding constraints; map inputs to references and specify exact strings, locations, and hierarchy.
- Ensure protocol-specific notes, such as endpoint domains, asynchronous handling, and region-specific caveats are respected.
When to use it
Use when productionizing Alibaba Cloud Model Studio image generation workflows, including multi-reference generation, character-consistent sets, typography/layout tasks, regional authentication, model selection, API integration, cost/rate-limit planning, troubleshooting, safety, rights, privacy, and output QA. Trigger the production contract phase before any paid call and follow the gated iteration process for quality assurance.
What it can touch
- Models and endpoints for Wan 2.7, Qwen-Image 2.0, and Z-Image Turbo as described in the guidance.
- Regions and workspace configurations, API keys stored in environment-like variables (DASHSCOPE_API_KEY).
- Endpoints for Singapore and China Bejing as listed, with correct Authorization header format and content type.
- Request structure using one user message, with image objects and a single text object; size specified as width*height.
Caveats
- Requires active verification of model IDs, regions, pricing, and quotas before deployment; live model lists may change.
- Do not place API keys in source, prompts, logs, patches, or artifacts; use region-appropriate, least-privilege keys.
- Be aware of potential asynchronous handling for Wan 2.7 long generation and edit tasks; persist IDs and statuses accordingly.
- The guidance emphasizes not changing chosen model, region, or prompt rewrite policy without explicit approval gates.
# Alibaba image models Treat this skill as the production guide for Alibaba Cloud Model Studio image inference, not for local Wan checkpoints, image-to-video, model training, or legacy one-purpose Wan editing endpoints. Check the live model list before every production integration because model IDs, aliases, regions, pricing, and quotas change. ## Establish the production contract Before generatin
What does the alibaba-image-models skill do?
Plan, prompt, call, edit, iterate, and productionize Alibaba Cloud Model Studio image generation with current Wan 2.7 Image, Qwen-Image 2.0, and Z-Image Turbo models. Use for Alibaba/DashScope text-to-image, multi-reference generation, instruction editing, character-consistent image sets, typography and layout work, regional authentication, model selection, API integration, cost/rate-limit planning, troubleshooting, safety, rights, privacy, and output QA.
How do I install it?
Run `npx skills add calesthio/generative-media-skills --skill alibaba-image-models --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From calesthio/generative-media-skills, a repository with 112 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.