Currently the LLM call blocks until the full response is generated. For large codebases, this can take 30-60 seconds with no feedback. Implement streaming via the Anthropic streaming API so users can see the LLM's analysis and changes being generated in real-time. Show a progress indicator or live output in the terminal.
Currently the LLM call blocks until the full response is generated. For large codebases, this can take 30-60 seconds with no feedback. Implement streaming via the Anthropic streaming API so users can see the LLM's analysis and changes being generated in real-time. Show a progress indicator or live output in the terminal.