[BUG] /context usage stuck at 0% breaks auto-compaction (+ output printed twice) (2.1.19)

Preflight Checklist

• [x] I have searched existing issues and this hasn't been reported yet
• [x] This is a single bug report (please file separate reports for different bugs)
• [x] I am using the latest version of Claude Code

What's Wrong?

On Claude Code v2.1.19, running /context prints the entire context usage output twice.

In addition to /context rendering twice, the total usage header is always 0%, despite category usage being non-zero. This appears to prevent auto-compaction from triggering, since compaction relies on the overall context usage threshold.

Additionally, the header shows the model usage as:

0k/200k tokens (0%)

even though the breakdown clearly shows non-zero token usage (System prompt, tools, messages, etc.).

❯ /context
  ⎿   Context Usage
     ⛁ ⛁ ⛁ ⛁ ⛁ ⛁ ⛁ ⛁ ⛁ ⛁   claude-opus-4.5 · 0k/200k tokens (0%)
     ⛁ ⛁ ⛁ ⛁ ⛁ ⛁ ⛁ ⛁ ⛁ ⛁
     ⛁ ⛁ ⛁ ⛁ ⛁ ⛁ ⛁ ⛁ ⛁ ⛁   Estimated usage by category
     ⛁ ⛁ ⛁ ⛁ ⛁ ⛁ ⛁ ⛁ ⛁ ⛁   ⛁ System prompt: 2.4k tokens (1.2%)
     ⛁ ⛁ ⛁ ⛁ ⛁ ⛶ ⛶ ⛶ ⛶ ⛶   ⛁ System tools: 15.5k tokens (7.7%)
     ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶   ⛁ Messages: 72.3k tokens (36.2%)
     ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶   ⛶ Free space: 65k (32.4%)
     ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛝ ⛝ ⛝   ⛝ Autocompact buffer: 45.0k tokens (22.5%)
     ⛝ ⛝ ⛝ ⛝ ⛝ ⛝ ⛝ ⛝ ⛝ ⛝
     ⛝ ⛝ ⛝ ⛝ ⛝ ⛝ ⛝ ⛝ ⛝ ⛝

     ⛁ ⛁ ⛁ ⛁ ⛁ ⛁ ⛁ ⛁ ⛁ ⛁   claude-opus-4.5 · 0k/200k tokens (0%)
     ⛁ ⛁ ⛁ ⛁ ⛁ ⛁ ⛁ ⛁ ⛁ ⛁
     ⛁ ⛁ ⛁ ⛁ ⛁ ⛁ ⛁ ⛁ ⛁ ⛁   Estimated usage by category
     ⛁ ⛁ ⛁ ⛁ ⛁ ⛁ ⛁ ⛁ ⛁ ⛁   ⛁ System prompt: 2.4k tokens (1.2%)
     ⛁ ⛁ ⛁ ⛁ ⛁ ⛶ ⛶ ⛶ ⛶ ⛶   ⛁ System tools: 15.5k tokens (7.7%)
     ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶   ⛁ Messages: 72.3k tokens (36.2%)
     ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶   ⛶ Free space: 65k (32.4%)
     ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛶ ⛝ ⛝ ⛝   ⛝ Autocompact buffer: 45.0k tokens (22.5%)
     ⛝ ⛝ ⛝ ⛝ ⛝ ⛝ ⛝ ⛝ ⛝ ⛝
     ⛝ ⛝ ⛝ ⛝ ⛝ ⛝ ⛝ ⛝ ⛝ ⛝

What Should Happen?

• /context should print the context usage only once
• Total usage should reflect the breakdown values, instead of showing 0%

Error Messages/Logs

Steps to Reproduce

1. Launch Claude Code
2. Write a couple of messages to fill up the context
3. Run /context
4. Observe that the output renders twice
5. Observe that total usage remains 0% despite non-zero category values

Claude Model

Opus

Is this a regression?

Yes, this worked in a previous version

Last Working Version

_No response_

Claude Code Version

2.1.19 (Claude Code)

Platform

Other

Operating System

Windows

Terminal/Shell

Windows Terminal

Additional Information

_No response_