Aliyun Qwen Omni

Use when tasks require all-in-one multimodal understanding or generation with Alibaba Cloud Model Studio Qwen Omni models, including image-plus-audio interac...

Security Scan

This skill is a small, instruction-only helper that only prepares a local JSON request payload for Alibaba's Qwen Omni models; it does not contact external services or request credentials.

Purpose & Capability

The name/description advertises full multimodal integration with Alibaba Cloud Qwen Omni, but the shipped files only produce a local example request payload and an agent prompt. No network calls, SDKs, or credential handling are included — so the skill is a lightweight payload/template helper rather than a full provider integration.

Instruction Scope

Runtime instructions only compile the included Python script and run it to write a JSON payload to disk. The instructions do not read other system files, environment variables, or transmit data externally.

Install Mechanism

There is no install specification; the skill is instruction-only plus a tiny script. Nothing is downloaded or written beyond the simple generated JSON file.

Credentials

The skill requests no environment variables, credentials, or config paths. Note: agents/openai.yaml contains a prompt string referencing "$alicloud-ai-multimodal-qwen-omni" but this is a prompt template, not an actual required env var.

Persistence & Privilege

The skill does not request always:true and contains no code that modifies agent/system configuration. It is user-invocable and can be run by the agent, which is normal.

Assessment

This skill is essentially a small template that writes an example request.json for Qwen Omni — it does not call Alibaba Cloud or handle credentials. If you expected a full integration that sends requests to Model Studio, note you'll need additional code and valid Alibaba Cloud credentials (and should verify any such code comes from a trusted source). It's safe to install from a security perspective, but its utility is limited unless you or the maintainer add an actual API client and credential handling. If you want a skill that calls the model service, ask the publisher for explicit instructions about which environment variables are required and how network requests are performed before providing any secrets.

Category: provider

Model Studio Qwen Omni

Validation

mkdir -p output/aliyun-qwen-omni
python -m py_compile skills/ai/multimodal/aliyun-qwen-omni/scripts/prepare_omni_request.py && echo "py_compile_ok" > output/aliyun-qwen-omni/validate.txt

Pass criteria: command exits 0 and output/aliyun-qwen-omni/validate.txt is generated.

Critical model names

Use one of these exact model strings:

• qwen3-omni-flash
• qwen3-omni-flash-realtime
• qwen-omni-turbo
• qwen-omni-turbo-realtime

Typical use

• Image + audio + text assistant
• Realtime multimodal agents
• Spoken responses grounded in visual input

Normalized interface (omni.chat)

Request

• model (string, optional): default qwen3-omni-flash
• text (string, optional)
• image (string, optional)
• audio (string, optional)
• response_modalities (array<string>, optional): e.g. ["text"], ["text","audio"]

Response

• text (string, optional)
• audio_url or audio_chunk (optional)
• usage (object, optional)

Quick start

python skills/ai/multimodal/aliyun-qwen-omni/scripts/prepare_omni_request.py \
  --output output/aliyun-qwen-omni/request.json

References

• references/sources.md