DashScope LLM
Security checks across static analysis, malware telemetry, and agentic risk
Overview
This is a simple, disclosed DashScope API wrapper; users should know it uses their DashScope API key and sends the prompt to Alibaba Cloud.
This skill appears benign and narrowly scoped. Before using it, set a dedicated DashScope API key, make sure you are comfortable sending the prompt text to DashScope, and use a trusted installed version of the OpenAI Python package.
Static analysis
No static analysis findings were reported for this release.
VirusTotal
VirusTotal findings are pending for this skill version.
Risk analysis
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
The request may consume the user's DashScope account quota or billing and uses whatever permissions that API key has.
The script reads a DashScope API credential from the environment. This is expected for calling DashScope, but users should notice that the registry metadata declares no required env vars or primary credential.
api_key=os.getenv("DASHSCOPE_API_KEY")Use a dedicated, least-privilege DashScope API key and avoid running it in environments where unintended keys may be present.
Any text placed in the prompt is transmitted to DashScope/Alibaba Cloud for processing.
The code sends the user-provided message to DashScope's external OpenAI-compatible API. This is disclosed and purpose-aligned, but it is still an external data flow.
base_url="https://dashscope.aliyuncs.com/compatible-mode/v1" ... messages=[{"role": "user", "content": args.message}]Do not send secrets, private documents, or regulated data unless the user's DashScope account and data-processing terms are appropriate.
Behavior depends on the OpenAI package version already installed in the environment.
The skill depends on an external Python package supplied by the user's environment, and the artifacts do not include a pinned dependency or install spec. There is no evidence of hidden installation or execution.
- the `openai` Python package installed
Install the OpenAI package from a trusted source and consider pinning a known-good version in the environment where this skill is used.
