Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add calesthio/generative-media-skills --skill alibaba-image-modelsgit clone --depth 1 https://github.com/calesthio/generative-media-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/calesthio/generative-media-skills/alibaba-image-models)<a href="https://agentmods.dev/skills/calesthio/generative-media-skills/alibaba-image-models"><img src="https://agentmods.dev/badge/skills/calesthio/generative-media-skills/alibaba-image-models/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/calesthio/generative-media-skills/alibaba-image-models"><img src="https://agentmods.dev/badge/skills/calesthio/generative-media-skills/alibaba-image-models.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00105 | $0.06478 |
| Opus 5 | $0.00053 | $0.03239 |
| Sonnet 5 | $0.00021 | $0.01296 |
| Haiku 4.5 | $0.00011 | $0.00648 |
Grade A, and why
alibaba-image-models scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 8d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
from urllib.parse import urljoin, urlsplit How it starts
The opening of the file, as written. The whole thing — 375 lines — stays where its author put it; the contents beside it link to each section on GitHub.
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 generating, record:
- deliverable and channel: hero, poster, product image, character sheet, storyboard, social crop, or edit;
- required pixel dimensions, aspect ratio, count, and file format;
- exact text, brand colors, protected areas, and layout hierarchy;
- reference-image roles and permission to use each reference;
- acceptable identity/product drift and whether exact typography must be composited later;
- target region, workspace, model ID or pinned snapshot, budget ceiling, and retention needs;
- approval gates: low-cost sample, selected direction, edit pass, final QA.
Announce the provider, model, region, number of requested images, and estimated charge before a paid call. Generate one or two low-cost candidates first. Do not silently change a chosen model, region, or prompt-rewrite policy.
Route deliberately
The following facts were verified against Alibaba Cloud documentation on 2026-07-09.
| Need | Prefer | Why and boundary |
|---|---|---|
| Broad generation, text rendering, brand palette, editing, up to nine references, bounding-box edits, or coherent sets | wan2.7-image-pro |
Most complete current Wan route. Pure text-to-image can reach the 4K specification; edits and image sets are limited to 2K. Standard mode returns 1–4 images; sequential mode can return up to 12. |
| Similar Wan workflows with faster/lower-cost output | wan2.7-image |
Same workflow family, but up to 2K. Validate quality on the actual brief. |
| Negative prompt or up to six variants in one synchronous request | qwen-image-2.0-pro |
Combined generation/editing route with 1–6 PNG outputs and a documented negative prompt. Editing accepts one to three images. |
| Faster Qwen generation/editing | qwen-image-2.0 |
Accelerated 2.0 route; retain the same sync protocol and test quality against Pro. |
| Fast, inexpensive generation-only portraits or product photos | z-image-turbo |
One PNG output, no editing, 512×512 through 2048×2048 range. Prompt rewriting doubles its documented per-image price. |
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 8d ago First seen · 375 lines · 105 tokens per session scan A 0f9344291909
alibaba-image-models is a skill published in the GitHub repository calesthio/generative-media-skills (170 stars, last pushed 1mo ago), licensed MIT. It adds 105 tokens to every session and 6,478 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
cliptalk-graphics-packager
Adds task-local graphic text layers such as title cards, top labels, product tags, price or parameter callouts, watermarks, and CTA overlays to an existing timeline.
image-authoring
Author images and diagrams as code — SVG, Pillow, Excalidraw, mermaid. Load when asked to draw, illustrate, or make an image, icon, logo, poster, or diagram.
bento-slides
Create and edit Bento presentations — self-contained .bento.html decks whose document is JSON. Use whenever the user wants a slide deck or presentation: from scratch, from source material, or by improving an existing file.
motion-graphics
Use when the user wants a short, design-led motion graphic where motion is the message: kinetic typography, stat or number count-up, chart/data-viz hit, logo sting, brand lockup, lower-third, callout, social overlay, animated headline/tweet/news item, motion poster, or quick captured-page highlight. Usually under 10s…
storyboard-generation
Rules and prompts for generating structured presentation slides and multi-panel storyboards.
excalidraw-visual-designer
Create and revise editable visual drawings directly in the Excalidraw website through the in-app browser. Use when the user asks Codex to draw in Excalidraw or update an Excalidraw canvas, including diagrams, flowcharts, architecture sketches, process maps, infographics, teaching visuals, product explanation graphics…