[FEATURE] Expose full context usage in statusline API

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

The statusline API currently only exposes conversation tokens (total_input_tokens + total_output_tokens), not the full context window usage including system overhead.

The context_window object passed to statusline scripts contains:
{
"total_input_tokens": 4589,
"total_output_tokens": 7738,
"context_window_size": 200000
}

This shows ~6% usage (12k/200k tokens).

Proposed Solution

Expected Behavior

The /context command shows the true breakdown:

• System prompt: 3.2k tokens
• System tools: 32.4k tokens
• MCP tools: 37.4k tokens
• Custom agents: 10.9k tokens
• Memory files: 3.0k tokens
• Messages: 1.2k tokens
• Total: 133k/200k (67%)

Feature Request

Please expose the full context usage in the statusline API, e.g.:
{
"total_input_tokens": 4589,
"total_output_tokens": 7738,
"context_window_size": 200000,
"total_used_tokens": 133000 // <-- new field
}

Use Case

I want to display a context usage progress bar in my statusline that accurately reflects how much context remains before needing to compact. Currently it shows ~6% when actual usage is ~67%, making it misleading.

Alternative Solutions

_No response_

Priority

Medium - Would be very helpful

Feature Category

CLI commands and flags

Use Case Example

_No response_

Additional Context

_No response_