Aliyun Qwen Text Embedding

Use when text embeddings are needed from Alibaba Cloud Model Studio models for semantic search, retrieval-augmented generation, clustering, or offline vector...

Security Scan

The skill is internally consistent with its stated purpose: it only prepares a local JSON embedding request for Alibaba Cloud Model Studio and requires no credentials or installs.

Purpose & Capability

Name, description, and included script align: the script builds a minimal payload for Model Studio embedding models and SKILL.md documents the correct model names and usage. The agent interface text references Alibaba Cloud embedding functionality (minor naming mismatch noted below) but otherwise matches purpose.

Instruction Scope

Runtime instructions only compile and run a small Python script to write a local request.json; they do not read unrelated files, access network endpoints, or reference credentials. The instructions stay within the stated task of preparing embedding requests.

Install Mechanism

No install spec is provided (instruction-only), so nothing is downloaded or installed. This minimizes disk/execute risk.

Credentials

The skill declares no required environment variables, no primary credential, and no config paths. That is proportionate because the included script only creates a local JSON payload and does not call external APIs.

Persistence & Privilege

No 'always' privilege is requested, the skill is user-invocable only, and it does not modify system or other skills' configuration. It writes only to its own output path when run.

Assessment

This skill is small and coherent: it only generates a local JSON request payload for Alibaba Cloud Model Studio embeddings. Before using in production, verify the source (unknown homepage) and the model names listed match the target API version you intend to call. The skill does not perform the API call itself, so you'll need to supply appropriate Alibaba Cloud credentials and a separate client/integration to send the generated request — review that integration for credential handling. Note the agents/openai.yaml default_prompt references "$alicloud-ai-search-text-embedding" which is a minor naming inconsistency (harmless but check prompts/variable names when wiring into agents).

Category: provider

Model Studio Text Embedding

Validation

mkdir -p output/aliyun-qwen-text-embedding
python -m py_compile skills/ai/search/aliyun-qwen-text-embedding/scripts/prepare_embedding_request.py && echo "py_compile_ok" > output/aliyun-qwen-text-embedding/validate.txt

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

Critical model names

Use one of these exact model strings as needed:

• text-embedding-v4
• text-embedding-v3
• text-embedding-v2
• text-embedding-v1
• qwen3-embedding-8b
• qwen3-embedding-4b
• qwen3-embedding-0.6b

Quick start

python skills/ai/search/aliyun-qwen-text-embedding/scripts/prepare_embedding_request.py \
  --text "Alibaba Cloud Model Studio" \
  --output output/aliyun-qwen-text-embedding/request.json

Notes

• Pair this skill with skills/ai/search/aliyun-dashvector-search/ or other vector-store skills.
• For image or multimodal embeddings, add dedicated multimodal embedding coverage separately.

References

• references/sources.md