Qwen Image Plus Sophnet

Generate images via Sophnet Qwen-Image-Plus and poll for task completion. Use when the user asks for Sophnet image generation, Qwen-Image-Plus, or requests an image from the Sophnet API.

Security Scan

The skill appears to do exactly what it says — call Sophnet's Qwen-Image-Plus API, poll for completion, and return image URLs — but there are small metadata and documentation mismatches you should be aware of before installing.

Purpose & Capability

The name/description align with the code: the included shell script posts a create-task to Sophnet and polls the task endpoint for results. However, registry metadata declares no required environment variables or binaries, while the SKILL.md and the script expect an API key (SOPHNET_API_KEY or --api-key) and the presence of curl. Also the docs show an absolute path rooted in a specific user's home (/home/shutongshan/...), which is a leftover developer artifact and not required for the skill to function.

Instruction Scope

Runtime instructions stay within the stated purpose — they build a JSON payload, call Sophnet endpoints, poll status, and print TASK_ID/STATUS/IMAGE_URL lines. The script does not attempt to read other files, system-wide configs, or unrelated environment variables. The notable issues: SKILL.md tells the user to export SOPHNET_API_KEY but that env var is not declared in metadata; the Quick Start uses a hardcoded absolute path to a specific user's workspace which is misleading.

Install Mechanism

There is no install specification (instruction-only with an included script). Nothing is downloaded or written to disk by an installer step. This is the lowest-risk install profile.

Credentials

The script requires a Sophnet API key (SOPHNET_API_KEY or --api-key), which is proportionate to an API client. The metadata, however, does not declare this required environment variable (inconsistency). The example output shows presigned OSS URLs that include query parameters like OSSAccessKeyId and Signature — those are typically temporary and come from the service, but such URLs embed credentials-like tokens in the URL, so be careful about sharing them publicly.

Persistence & Privilege

The skill does not request persistent/always-on presence (always:false) and does not modify other skills or system configs. It only runs the supplied script when invoked.

Assessment

This skill is internally consistent with its stated purpose (calling Sophnet's image API). Before installing: 1) Be prepared to supply a Sophnet API key (SOPHNET_API_KEY or pass --api-key) and ensure the key's scope/quota are appropriate. 2) Confirm curl is available where the agent runs (the metadata omitted this requirement). 3) Ignore the example's hardcoded /home/... path — run the script from the skill location or adjust the command to the actual install path. 4) Note that returned IMAGE_URLs may be presigned links containing temporary tokens; avoid posting those URLs in public if you want to keep them private. 5) If you need stronger assurance, review or run the included scripts in an isolated environment first.

Qwen-Image-Plus (Sophnet) Image Generation

Use the Sophnet image generator API to create an image task, poll until it finishes, then return the image URL.

Quick Start

Set the API key (preferred):

export SOPHNET_API_KEY="YOUR_API_KEY"

Run the script with an absolute path (do NOT cd to the skill directory):

bash /home/shutongshan/.openclaw/workspace/skills/qwen-image-plus-sophnet/scripts/generate_image.sh --prompt "your prompt"

Script Options

• --prompt (required): user prompt
• --negative-prompt (optional)
• --size (optional, default 1024*1024)
• --n (optional, default 1)
• --watermark (optional, default false)
• --prompt-extend (optional, default true)
• --api-key (optional, overrides SOPHNET_API_KEY)
• --poll-interval (optional, default 2)
• --max-wait (optional, default 300)

Output Contract

The script prints:

• TASK_ID=...
• STATUS=succeeded
• IMAGE_URL=... (one or more lines)

Use the IMAGE_URL value to respond to the user.

Workflow

1. POST create-task with model=Qwen-Image-Plus and user prompt
2. Poll GET task status until SUCCEEDED
3. Extract url and return to the user

Real Example (captured run)

Prompt:

A scenic mountain landscape in ink wash style

Command:

bash /home/shutongshan/.openclaw/workspace/skills/qwen-image-plus-sophnet/scripts/generate_image.sh \
  --prompt "A scenic mountain landscape in ink wash style" \
  --negative-prompt "blurry, low quality" \
  --size "1024*1024" \
  --n 1 \
  --watermark false \
  --prompt-extend true

Output:

TASK_ID=7BWFICt0zgLvuaTKg8ZoDg
STATUS=succeeded
IMAGE_URL=https://dashscope-result-wlcb-acdr-1.oss-cn-wulanchabu-acdr-1.aliyuncs.com/7d/d5/20260203/cfc32567/f0e3ac18-31f6-4a1a-b680-a71d3e6bcbe03032414431.png?Expires=1770714400&OSSAccessKeyId=LTAI5tKPD3TMqf2Lna1fASuh&Signature=fF12GZ7RgGsC7OpEkGCapkBUXws%3D

Common Errors

• Error: No API key provided. -> set SOPHNET_API_KEY or pass --api-key
• STATUS=failed -> check key permissions/quota or prompt parameters
• Error: url not found in response -> inspect API response manually