Anthropic has expanded its Claude AI assistant from a conversational chatbot into a no-code application platform. A new beta upgrade to the Artifacts workspace, released on 25 June, lets users build, host and share fully-functional AI-powered web apps directly inside Claude. The feature is available across the service’s Free, Pro and Max tiers, with every app’s processing billed to the viewer’s own Claude account rather than the creator’s. The company says users have already produced more than 500 million artifacts—ranging from educational games to data-analysis tools—since the workspace debuted last year. By embedding Claude’s intelligence into those creations, the upgrade eliminates the need to copy results into external software and aligns with the industry trend dubbed “vibe coding,” where natural-language prompts replace conventional programming. Anthropic is positioning the move as a counter to OpenAI’s Canvas interface while broadening Claude’s ecosystem. On 8 July, the firm previewed a GitHub app that lets developers call “Claude Code” inside repositories through a simple slash command, keeping project data on a customer’s own infrastructure. The San Francisco-based start-up is using a freemium model to attract new customers and says the platform applies the same multi-layered content-moderation and safety policies that govern its language models. Anthropic expects the combination of community sharing, no-code creation and professional integrations to deepen user engagement and create a new source of competitive differentiation in the crowded generative-AI market.