window

Compute current context-window usage from session JSONL logs.

Updated Feb 22, 2026
One-click install
npx skills add https://github.com/mbriggsy/ai-learning-journey --skill window
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: window
Source: https://github.com/mbriggsy/ai-learning-journey/tree/main/projects/skills/window
Command: npx skills add https://github.com/mbriggsy/ai-learning-journey --skill window

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Reports the current context-window usage in a concise, phone-friendly line.

Core Features & Use Cases

  • Tracks and reports the active session's context usage, delivering a single line suitable for small screens.
  • Works via /window or natural language prompts about remaining context, and reads the live logs to compute usage.
  • Use cases include quickly assessing whether to continue a session before hitting the 1M window.

Quick Start

Type /window to retrieve the current context-window usage as a single line.

Frequently Asked Questions about window

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I check remaining context window usage on a mobile phone?

Check context window usage by invoking the /window command, which aggregates active session JSONL logs and token counts to compute and emit a compact, phone-friendly status line reporting live usage.

What is a context window and when do I need to monitor it?

A context window is the maximum token capacity a session supports. Monitor context window usage to assess whether to continue a session before hitting the 1M token limit, preventing unexpected truncation.

Does the context window check work on small screens without overwhelming the display?

Yes, the context window check works on small screens by design. It computes live token usage and delivers the result as a single line, ensuring the output remains concise and readable on mobile phones.

How to compute live context token counts from session logs?

Compute live context token counts from session logs by triggering the /window command, which aggregates the active session's JSONL logs to calculate current token usage and reports it as a single status line.

When should I avoid relying on a single-line context summary?

Avoid relying on a single-line context summary when you need detailed token breakdowns or historical usage trends, as this approach intentionally provides only a compact, live status line for quick checks on mobile screens.