[FEATURE] Add YOLO mode toggle support to Claude Code CLI

Preflight Checklist

• [x] I have searched existing requests and this feature hasn't been requested yet
• [x] This is a single feature request (not multiple features)

Problem Statement

When working with Claude Code CLI, there are situations where users want to quickly execute commands without strict confirmations or safety checks - what I call "YOLO mode" (You Only Live Once).

Currently, the CLI always requires confirmations for potentially risky operations, which can slow down workflows when users are confident about their changes.

I would like to request the addition of a YOLO mode toggle that allows users to:

1. Bypass confirmation prompts for file modifications
2. Automatically apply suggested changes without manual review
3. Proceed with risky operations (like git operations, file deletions, etc.) without interruption

This should be:

• Explicitly opted-in by users (not default behavior)
• Clearly documented with warnings about the risks
• Available as a CLI flag (e.g., --yolo or --force-mode)
• Optionally configurable in the config file

Proposed Solution

The solution should include:

1. CLI Flag: Add a --yolo or --force flag to the Claude Code CLI that disables confirmation prompts

Example: claude --yolo "modify all files as needed"

1. Configuration Option: Allow users to set yolo_mode: true in their .claude/config.json file

1. Explicit Warnings: When YOLO mode is activated, display a clear warning message that this mode bypasses safety checks

1. Behavior Changes:

• File modifications don't require "yes/no" confirmation
• Git operations execute without prompts
• Batch operations proceed without pause points
• Risk-related warnings are still displayed but don't block execution

1. Implementation Details:

• Add a --yolo flag to the CLI argument parser
• Add a yolo_mode boolean to the config structure
• Pass this flag through to the decision-making functions
• Modify confirmation logic to skip interactive prompts when enabled

Alternative Solutions

Current workarounds users might be using:

• Manually confirming each action and accepting all changes (slow and tedious)
• Using shell scripts to pipe responses into Claude Code (unreliable)
• Running Claude Code in separate windows for different tasks (context switching)
• Disabling the tool entirely for time-sensitive operations (loses functionality)

A dedicated YOLO mode would be cleaner and more maintainable than these workarounds.

Priority

Critical - Blocking my work

Feature Category

CLI commands and flags

Use Case Example

As a developer working on a large codebase, I often need to make multiple bulk changes across many files. My typical workflow is:

1. I'm working on refactoring code from TypeScript to use a new design pattern
2. I need Claude Code to identify all files that need modification
3. Claude Code proposes changes to 50+ files automatically
4. Currently, I need to manually confirm each change one by one, which takes hours

With YOLO mode enabled, I could run: claude --yolo "refactor all these files to use the new pattern" and let Claude Code process all changes without confirmations, significantly speeding up my workflow. This would save me from repetitive clicking while still allowing me to review the changes afterward in git before committing.

Additional Context

_No response_