compact-guide

Summarize and prune conversation context to stay within token budgets.

125|35|Updated Jan 21, 2026
One-click install
npx skills add https://github.com/jh941213/my-claude-code-asset --skill compact-guide
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: compact-guide
Source: https://github.com/jh941213/my-claude-code-asset/tree/main/skills/compact-guide
Command: npx skills add https://github.com/jh941213/my-claude-code-asset --skill compact-guide

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps maintain high-quality conversations by automatically summarizing and pruning context to stay within token budgets, preventing drift and confusion over long sessions.

Core Features & Use Cases

  • Adaptive summarization: periodically compress recent dialogue to preserve essential info.
  • Context pruning: remove outdated or redundant details while keeping the thread coherent.
  • Use Case: In a long brainstorming session, apply /compact every few steps to keep the context lean, then use /handoff to save a clean HANDOFF.md.

Quick Start

Start a new session and run /compact after every 3-5 tasks to compress the context, then run /clear before switching topics.

Frequently Asked Questions about compact-guide

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

FAQPage Schema
How do I summarize and prune conversation context to stay within token limits?

To manage context within token limits, periodically compress recent dialogue and remove redundant details using the /compact command. This adaptive summarization preserves essential information while keeping the chat thread lean and coherent.

What is the best way to handle conversation handoffs when switching topics?

The best way to handle conversation handoffs is to run /handoff to save a clean HANDOFF.md file, then use /clear to reset the context before switching topics, ensuring no drift or confusion in the new session.

When do I need context compression for long brainstorming sessions?

You need context compression for long brainstorming sessions when the dialogue approaches token limits. Apply /compact every few steps or 3-5 tasks to remove outdated details and keep the context lean without losing the thread's coherence.

Does adaptive summarization work for ongoing tutoring scenarios?

Adaptive summarization works for ongoing tutoring scenarios by periodically compressing recent dialogue to preserve essential info. It manages long context by pruning outdated details while keeping the educational thread coherent.

Why does my chat lose coherence during long sessions despite staying under the token limit?

Your chat loses coherence because outdated or redundant details accumulate over long sessions. Applying context pruning with /compact removes this noise, preserving essential info and maintaining high-quality conversations within token budgets.