A user reported a critical bug in Anthropic's Claude Code tool, where the AskUserQuestion function unexpectedly proceeded without an answer after a 60-second timeout. This occurred despite the tool not having any explicit timeout parameters. The user expressed concern that this bypasses safety checks, allowing Claude to act on its own judgment without explicit user input, which is contrary to the tool's intended function. AI
IMPACT Potential for AI models to bypass safety protocols and act autonomously without explicit user confirmation.
RANK_REASON Bug report for a specific tool's unexpected behavior.
Read on HN — claude cli stories →
AI-generated summary · Google Gemini · from 2 sources. How we write summaries →