[Feature Request] Add code change review interface with accept/skip options for individual modifications

Found 3 possible duplicate issues:

1. https://github.com/anthropics/claude-code/issues/4322
2. https://github.com/anthropics/claude-code/issues/2502
3. https://github.com/anthropics/claude-code/issues/7760

This issue will be automatically closed as a duplicate in 3 days.

• If your issue is a duplicate, please close it and 👍 the existing issue instead
• To prevent auto-closure, add a comment or 👎 this comment

🤖 Generated with Claude Code

closing as this issue matches what i want:

https://github.com/anthropics/claude-code/issues/2502

Extended Thinking Configuration - Partial Support Exists

Extended thinking can already be controlled, but options are limited.

Current Configuration

// ~/.claude/config.json
{
  "alwaysThinkingEnabled": false  // Disable extended thinking
}

This works but is all-or-nothing. No granular control.

Missing Configuration Options

What users want:

{
  "extendedThinking": {
    "enabled": true,
    "budget": "medium",  // "low" | "medium" | "high" | "unlimited"
    "showThinkingProcess": true,  // Show <thinking> tags
    "maxThinkingTokens": 10000,   // Cap thinking length
    "autoCollapse": true          // Collapse thinking in UI
  }
}

Implementation Approach

Map budget to API parameters:

const THINKING_BUDGETS = {
  low: { max_tokens: 5000, temperature: 0.7 },
  medium: { max_tokens: 10000, temperature: 1.0 },
  high: { max_tokens: 20000, temperature: 1.0 },
  unlimited: { max_tokens: null, temperature: 1.0 }
};

function createAPIRequest(config: ExtendedThinkingConfig) {
  const budget = THINKING_BUDGETS[config.budget];
  
  return {
    model: 'claude-opus-4-5',
    extended_thinking: {
      enabled: config.enabled,
      max_thinking_tokens: budget.max_tokens
    },
    temperature: budget.temperature
  };
}

UI for thinking blocks:

// If config.autoCollapse = true:
<details>
  <summary>🤔 Thinking (2,453 tokens)</summary>
  <thinking>
    ... thinking content ...
  </thinking>
</details>

// If config.showThinkingProcess = false:
// Don't render thinking blocks at all (save UI space)

Why This is Tricky

Extended thinking is model-level, not config-level. The API controls it, not Claude Code.

What Claude Code CAN control:

• Whether to send extended_thinking: true to API
• How to render <thinking> blocks in UI
• Token budget requests (but API may override)

What it CANNOT control:

• Actual thinking behavior (that's model-side)
• Thinking quality/depth
• When model decides to think

Workaround: Model Selection

Use different models for different thinking needs:

{
  "profiles": {
    "fast": {
      "model": "claude-sonnet-4",
      "extended_thinking": false
    },
    "deep": {
      "model": "claude-opus-4-5",
      "extended_thinking": true
    }
  }
}

Then: claude --profile deep

Feature Request for Anthropic API

To make this work, API needs:

{
  "extended_thinking": {
    "enabled": true,
    "thinking_budget": "medium",  // Let users control depth
    "show_process": false         // Don't return <thinking> blocks
  }
}

---

Verdict: Partially exists, but limited by API capabilities.
Recommend: Request API-level thinking controls from Anthropic.

This issue has been automatically locked since it was closed and has not had any activity for 7 days. If you're experiencing a similar issue, please file a new issue and reference this one if it's relevant.