[BUG] claude -m sonnet -p "plan this project" style prompt usually hanging on an error

Environment

• Platform (select one):
• [ ] Anthropic API
• [ ] AWS Bedrock
• [ ] Google Vertex AI
• [ x ] Other: Claude Max
• Claude CLI version: 1.0.16
• Operating System: macOS 15.4.1
• Terminal: iTerm2

Bug Description

I have a script that calls claude-p to plan the next task, then again to do dev, then a third time to review the code. The first part, the planner, has been hanging most of the time lately (not always) on a particular error. I don't know what it's doing at that time; it could be running a build or something?

==================================================================
This error originated either by throwing inside of an async function without a catch block, or by rejecting a promise which was not handled with .catch(). The promise rejected with the reason:
Error: Expected content to be an assistant message, but got "user"
at yS2 (file:///opt/homebrew/lib/node_modules/@anthropic-ai/claude-code/cli.js:2342:9744)
at process.processTicksAndRejections (node:internal/process/task_queues:105:5)
at async bE5 (file:///opt/homebrew/lib/node_modules/@anthropic-ai/claude-code/cli.js:2356:8265)

Steps to Reproduce

1. run claude -m sonnet -p 'do planning'
2. try again if it doesn't fail
3. <!-- And so on... -->

Expected Behavior

do work

Actual Behavior

error displayed, and then it never completes

Additional Context