This is a minimal test harness for validating Alibaba Cloud's QVQ multimodal AI models, which handle visual reasoning tasks. It's not a production tool but rather a developer utility that confirms the main QVQ skill is set up correctly with the right models and payload structure. You run a Python script that generates a sample request JSON, basically a sanity check before you start building with the actual multimodal skill. If you're integrating AliCloud's vision models into Claude workflows, this gives you a quick way to verify everything's wired up. It's passed security audits from three providers and has seen modest adoption at 120 installs, though as pure test infrastructure that's probably appropriate.
npx -y skills add cinience/alicloud-skills --skill alicloud-ai-multimodal-qvq-test --agent claude-codeInstalls into .claude/skills of the current project.
Select a file.
wshobson/agents
dbt-labs/dbt-agent-skills
github/awesome-copilot