Individual skill

qwen-image-gen

0upvotes

为 Qwen-Image 编写和改写生图提示词，区分局部编辑与参考主体创作，处理同人物系列、画幅及原图验收；用户要求生成时连接其已有的 Qwen 图像工作流。不用于下载权重、通用模型运维或其他图像模型。

skills/qwen-image-gen/SKILL.md

Read pinned source

Source commit 01c8ffe0bc94cfe1dbfc7e1de537399253d04f39

What was checked

Matches the observed repository HEAD: Yes
Scanner rules are current: Yes
Repository scan complete: No
Skill scan complete: Yes
Observed behavior: Local scripts
Last checked: Nov 11, 2026

These are observations at the recorded commit, not a guarantee of safety. Review the source before running it.

Install this skill

Choose an agent in your Stack or copy a command below. Commands fetch the pinned commit and copy only this skill directory. Existing destinations are refused. Copying a command does not install anything.

An install command is unavailable: current rules, complete repository and skill scans, an eligible behavior level and a permitted license are required.

Repository context

Repository stars (not skill ratings): 241
Repository last updated: Sep 22, 2026
License: Apache-2.0

Observed agent integration: No specific integration evidence. Standard SKILL.md installation does not imply verified compatibility.

agent-skills Repository topic
ai-agents Repository topic
codex Repository topic
design Repository topic
evals Repository topic
engineering Derived category

Files and excerpt

File metadata and permitted, redacted excerpts come from the recorded scan. Source links are pinned to that commit.

Browse files

ASSET-LICENSE.mddoc · 1,402 bytes
LICENSEasset · 1,066 bytes
README.en.mddoc · 5,761 bytes
README.mddoc · 5,950 bytes
SKILL.mdcore · 4,347 bytes
agents/openai.yamlasset · 325 bytes
assets/examples/ab/README.mddoc · 1,573 bytes
assets/examples/ab/baker-A.pngasset · 1,417,846 bytes
assets/examples/ab/baker-B.pngasset · 1,362,695 bytes
assets/examples/ab/baker-reference.pngasset · 5,132,293 bytes
assets/examples/ab/cases.jsonasset · 13,810 bytes
assets/examples/ab/checksums.jsonasset · 2,163 bytes
assets/examples/ab/poster-A.pngasset · 4,159,596 bytes
assets/examples/ab/poster-B.pngasset · 3,275,985 bytes
assets/examples/ab/reach-A.pngasset · 1,460,413 bytes
assets/examples/ab/reach-B.pngasset · 1,530,603 bytes
assets/examples/ab/stool-A.pngasset · 1,644,728 bytes
assets/examples/ab/stool-B.pngasset · 1,595,159 bytes
assets/examples/portraits/README.mddoc · 612 bytes
assets/examples/portraits/manifest.jsonasset · 3,622 bytes
assets/examples/portraits/noir-portrait.pngasset · 2,980,111 bytes
assets/examples/portraits/noir-portrait.txtdoc · 704 bytes
assets/examples/portraits/rooftop-green.pngasset · 2,987,303 bytes
assets/examples/portraits/rooftop-green.txtdoc · 755 bytes
assets/examples/portraits/strawberry-cake.pngasset · 2,994,912 bytes
assets/examples/portraits/strawberry-cake.txtdoc · 753 bytes
assets/manifest.jsonasset · 4,625 bytes
references/integration.mddoc · 1,951 bytes
references/rewriting.mddoc · 6,651 bytes
references/shared-runtime.mddoc · 5,756 bytes
references/tested-behavior.mddoc · 3,462 bytes
references/workbench-adapter.mddoc · 1,756 bytes
runtime/core.mjsruns · 11,722 bytes
runtime/manifest.jsonasset · 185 bytes
scripts/prepare_workbench_request.pyruns · 6,848 bytes
tests/test_prepare_request.pyruns · 5,920 bytes
tests/test_runtime.mjsruns · 9,255 bytes

Read excerpt


# Qwen Image Gen

将创作需求转成可执行的提示词、参考图关系和尺寸决策。默认由当前 Agent 改写，无需额外部署提示词增强模型。参考图任务需要视觉理解能力；纯文本执行者应先取得准确的图片描述。

## 先区分任务

- **提示词、改写、翻译**：交付可复制文本与尺寸建议，不提交生图。
- **文生图**：描述最终可见画面。身份、数量、文字、动作、颜色和位置是固定要求；只补全必要的构图与光照细节。
- **修改这张图**：先说改哪里、改成什么，再简洁锁定未修改内容。改变姿势时不能同时要求姿势不变。
- **参考主体创作新图**：先实际看图，明确哪张图提供身份、服装或背景，再描述新姿势与场景。不要只靠重复人物形容词维持同一张脸。

写提示词前读 [rewriting.md](references/rewriting.md)。保留用户的创作意图，不用通用美化词覆盖具体要求。简单需求不必扩成长文。

## 把单张、系列和画幅拆开