Anthropic hides operational details of Claude Code

Anthropic has recently updated Claude Code, its AI coding tool, modifying the progress display. The most significant change concerns hiding the names of the files the tool is operating on, whether reading, writing, or modifying.

Developer reaction

This choice has provoked a strong negative reaction from the developer community. Many argue that the visibility of the files involved is fundamental to understanding the AI's workflow, for debugging, and for verifying the correctness of operations. The lack of this information makes it more difficult to use the tool effectively and increases the risk of errors.

For those evaluating on-premise deployments, there are trade-offs between the control and visibility offered by self-hosted solutions and the ease of use of cloud services. AI-RADAR offers analytical frameworks on /llm-onpremise to evaluate these aspects.