On Monday, Anthropic unveiled a web application for its popular AI coding assistant, Claude Code. This innovative platform allows developers to create and manage multiple AI coding agents directly from their browsers.
The web version of Claude Code is currently being rolled out to subscribers of Anthropic’s $20-per-month Pro plan, as well as its higher-tier Max plans priced at $100 and $200 per month. Users on these plans can easily access Claude Code by visiting claude.ai— the same site for Anthropic’s consumer chatbot—and clicking on the “Code” tab, or they can utilize the Claude iOS app.
This launch reflects Anthropic’s ongoing efforts to enhance Claude Code’s capabilities, evolving it from a command-line interface (CLI) tool, traditionally used by developers, into a more versatile web-based platform. By moving Claude Code online, Anthropic aims to provide developers with the flexibility to create AI coding agents in diverse settings.
The competition in the AI coding tools space is intensifying. Once dominated by Microsoft’s GitHub Copilot, the landscape now includes formidable players like Cursor, Google, OpenAI, and Anthropic, each offering robust AI coding solutions—many of which are already accessible via the web. Nonetheless, Claude Code has emerged as one of the frontrunners, with user growth expanding tenfold since its broader debut in May. It now contributes over $500 million to Anthropic’s annual revenue.
In an exclusive interview with TechCrunch, Anthropic’s product manager Cat Wu highlighted that the success of Claude Code can be attributed largely to the company’s advanced AI models, which have gained favor among developers in recent years. Wu also mentioned that the Claude Code team strives to incorporate elements of enjoyment into the product wherever possible.
Looking ahead, Wu stated that Anthropic plans to further integrate Claude Code into various environments, although the terminal will likely remain its primary base. “Our key focus as we move forward is to ensure that the CLI product remains the most intelligent and customizable option for utilizing coding agents,” said Wu. “However, we’re committed to making Claude Code accessible everywhere, with web and mobile being significant milestones in this journey.”
Techcrunch event
San Francisco
|
October 27-29, 2025
Anthropic claims that an impressive 90% of the Claude Code product itself is generated by the company’s AI models. Wu, who previously held engineering roles, noted that she rarely types at a keyboard these days, primarily focusing on reviewing the outputs generated by Claude Code.
Early AI coding tools functioned like autocomplete features, completing lines of code as developers typed. However, the more autonomous nature of AI coding tools like Claude Code enables developers to set up agents that operate independently. This evolution has transformed millions of software engineers into managers of AI coding assistants in their everyday work routines.
Yet, not all developers welcome this change. A recent study revealed that some engineers performed slower with AI coding tools like Cursor. Researchers speculated that this lag may be due to the time spent on prompting the AI and waiting for its responses, detracting from their focus on other tasks. Additionally, AI coding tools can struggle with large, complex codebases, leading engineers to invest significant time rectifying errors from the AI models.
Nevertheless, Anthropic is determined to advance the development of AI coding agents. CEO Dario Amodei recently predicted that AI could soon generate 90% of the code for software engineers. While this may hold true within Anthropic, its widespread adoption across the industry may take some time.
