← Back to Skills Marketplace
qwencloud-image-generation
by
Cuixiaoyang123
· GitHub ↗
· v0.2.0
· MIT-0
260
Downloads
0
Stars
1
Active Installs
2
Versions
Install in OpenClaw
/install qwencloud-image-generation
Description
[QwenCloud] Generate and edit images using Wan and Qwen Image models. Supports text-to-image, image editing (style transfer, subject consistency, text render...
Usage Guidance
Before installing or invoking this skill: 1) Expect to provide a QwenCloud/DashScope API key (DASHSCOPE_API_KEY) — the skill will not work without it even though the manifest lists none. 2) Review the included scripts (qwencloud_lib.py, image_lib.py, gossamer.py, image.py) because they load .env files from the current directory or repo root and may write files under the repo (e.g., .agents/state.json) or suggest appending to agent config files; run them only if you trust the repository. 3) Do not paste your real API key into chat; follow the SKILL.md advice to use a .env with a placeholder and set the real key in your environment. 4) If you need stronger isolation, run the scripts in a throwaway/VM/container or inspect/execute them manually rather than allowing an agent to run them automatically. 5) Ask the skill author or maintainer to update the registry metadata to declare DASHSCOPE_API_KEY and any other required env vars and to clearly document which files the skill will read or modify — that will resolve the primary incoherence noted here.
Capability Analysis
Type: OpenClaw Skill
Name: qwencloud-image-generation
Version: 0.2.0
The skill bundle is a legitimate tool for generating and editing images via the QwenCloud/DashScope API. It contains well-structured Python scripts (image.py, qwencloud_lib.py) that handle API authentication, image processing, and local file uploads to Alibaba Cloud OSS as required for the service. The SKILL.md includes proactive security instructions directing the AI agent to never reveal API keys in plaintext. While it includes an update-check mechanism (gossamer.py) that prompts for the installation of a sibling skill via 'npx skills add', this behavior is consistent with the OpenClaw ecosystem's dependency management and lacks malicious intent.
Capability Tags
Capability Assessment
Purpose & Capability
The skill's name/description match an image-generation integration, and the code implements calls to a DashScope/QwenCloud API. However the registry metadata claims no required environment variables or primary credential while the code and SKILL.md clearly expect a DASHSCOPE_API_KEY (and optionally QWEN_BASE_URL / QWEN_REGION). That mismatch (manifest says 'none' but runtime needs an API key) is incoherent and should be resolved.
Instruction Scope
SKILL.md and the included scripts instruct the agent to load local .env files, search repository paths, resolve and upload local image files, poll remote tasks, and (via references/agent-compatibility.md) read and optionally append to agent config files (CLAUDE.md, AGENTS.md, etc.). The bundled gossamer.py also reads/writes repo state files and may run subprocesses. These file reads/writes and subprocess calls go beyond a simple 'call remote API' helper and were not reflected in the manifest.
Install Mechanism
There is no external install spec — this is an instruction+script bundle included in the skill. No remote downloads or archive extraction are performed by the skill itself. The only external install recommendation is user-facing guidance to run npx to add an update-check skill (user-initiated, not automatic).
Credentials
The manifest declares no required env vars, but the code and documentation expect DASHSCOPE_API_KEY (primary API key) and also honor QWEN_BASE_URL and QWEN_REGION. The libraries load .env files into the environment (load_dotenv), which can import arbitrary environment variables from repo/nested .env files. Requesting and accessing an API key is proportionate for this skill, but the manifest omission and the general .env loading behavior (which can expose other secrets present in repo files) are concerning.
Persistence & Privilege
The skill is not always-enabled and does not demand permanent autonomous inclusion (always:false). However scripts (agent-compatibility.md and gossamer.py) include code that can write to repository config/state files (.agents/state.json, project config markers) and suggest installing update-check via npx. Those behaviors are not automatically executed but — if run — will modify local repo state. SKILL.md does state to ask the user before modifying files.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install qwencloud-image-generation - After installation, invoke the skill by name or use
/qwencloud-image-generation - Provide required inputs per the skill's parameter spec and get structured output
Version History
v0.2.0
QwenCloud Image Generation Skill v0.2.0
- Added support for new Wan models: wan2.7-image and wan2.7-image-pro (including 4K generation, expanded editing capabilities, and more reference images).
- Updated mode/model selection guide to recommend wan2.7-image(-pro) for advanced and higher-quality text-to-image and image editing.
- Clarified usage for pure text-to-image and image editing scenarios, highlighting the latest model options.
- Updated model tables to include feature distinctions (e.g. 4K, interactive editing, sequential multi-image).
- General documentation improvements and reorganization for clearer guidance.
- No functional code changes; documentation only.
v0.1.0
Initial release of qwencloud-image-generation.
- Supports image generation and editing using Wan and Qwen Image models: text-to-image, style transfer, subject consistency, and text rendering.
- Provides detailed model and mode selection guides for different use cases.
- Includes clear guidance on API key setup, security best practices, and agent compatibility.
- Supports both synchronous and asynchronous execution, multi-image input, and interleaved text-image output.
- Warns and prevents use of incompatible Coding Plan keys.
Metadata
Frequently Asked Questions
What is qwencloud-image-generation?
[QwenCloud] Generate and edit images using Wan and Qwen Image models. Supports text-to-image, image editing (style transfer, subject consistency, text render... It is an AI Agent Skill for Claude Code / OpenClaw, with 260 downloads so far.
How do I install qwencloud-image-generation?
Run "/install qwencloud-image-generation" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is qwencloud-image-generation free?
Yes, qwencloud-image-generation is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does qwencloud-image-generation support?
qwencloud-image-generation is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created qwencloud-image-generation?
It is built and maintained by Cuixiaoyang123 (@cuixiaoyang123); the current version is v0.2.0.
More Skills