On the finish of February, Anthropic introduced Claude Code. Within the eight months since then, the coding agent has arguably turn out to be the corporate’s most vital product, serving to it carve out a distinct segment for itself within the extremely aggressive AI market. Now, Anthropic is making it simpler for builders to make use of Claude Code in additional locations with a new internet interface for accessing the agent.
To get began, you may want join Claude to your GitHub repositories. From there, the method of utilizing the agent is identical as if it had direct terminal entry. Describe what you want from it, and the agent will take it from there. Claude will present progress updates whereas it really works, and you’ll even steer it in real-time with extra prompts. By means of the net interface, it is also potential to assign Claude a number of coding duties to run in parallel.
“Each Claude Code activity runs in an remoted sandbox setting with community and filesystem restrictions. Git interactions are dealt with by a safe proxy service that ensures Claude can solely entry licensed repositories — serving to maintain your code and credentials protected all through all the workflow,” stated Anthropic.
Along with making Claude Code obtainable on the net, Anthropic is releasing a preview of the agent inside its iOS app. The corporate warns the combination is early, and that it hopes “to shortly refine the cellular expertise primarily based in your suggestions.”
Professional and Max customers can begin utilizing Claude Code on the net at present. Anthropic notes any cloud periods share the identical price limits with all different Claude Code utilization.