How does the skill know when to suggest a compact command?

The skill tracks 'Edit' and 'Write' tool calls. By default, it suggests a compaction after 50 calls, providing periodic reminders every 25 calls thereafter to keep your session lean.

Why use manual compaction instead of automatic?

Automatic compaction triggers at arbitrary points, often in the middle of a task where it might discard necessary context. Strategic manual compaction allows you to clear history only at logical transition points.

Does this skill execute the /compact command automatically?

No, it acts as a strategic advisor. It notifies you when it is a good time to compact, but leaves the final decision and execution to the user.

Can I customize the suggestion frequency?

Yes, you can adjust the COMPACT_THRESHOLD environment variable to change how many tool calls occur before the first suggestion is made.

Strategic Compact for Claude Code

Name: Strategic Compact for Claude Code
Author: affaan-m

byaffaan-m

•

112,917

•

生产力与工作流

Manages context window efficiency by suggesting manual compaction at logical task boundaries to prevent critical context loss.

Strategic Compact is a performance optimization skill for Claude Code that replaces unpredictable automatic compaction with intelligent, manual reminders. By monitoring tool usage through PreToolUse hooks, it identifies optimal moments—such as after a research phase or following a successful debug—to suggest the /compact command. This ensures that essential implementation details are preserved while unnecessary session history is cleared, maintaining high model performance and reducing token overhead during complex development workflows.

主要功能

01112,917 GitHub stars

02Prevents context loss during high-stakes multi-step implementations

03Identifies logical task boundaries for safe context reduction

04Integrates directly with Claude Code's PreToolUse hook system

05Provides configurable thresholds for initial and recurring reminders

06Monitors tool usage counts for Edit and Write operations

使用场景

01Resetting session history after a complex debugging cycle to start the next task with a clean slate

02Clearing research context after a plan has been finalized and implementation is ready to begin

03Managing token usage in long-form development sessions without losing track of current file changes

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add affaan-m/everything-claude-code strategic-compact

For use in Claude.ai and ChatGPT

主要功能

01112,917 GitHub stars

02Prevents context loss during high-stakes multi-step implementations

03Identifies logical task boundaries for safe context reduction

04Integrates directly with Claude Code's PreToolUse hook system

05Provides configurable thresholds for initial and recurring reminders

06Monitors tool usage counts for Edit and Write operations

使用场景

01Resetting session history after a complex debugging cycle to start the next task with a clean slate

02Clearing research context after a plan has been finalized and implementation is ready to begin

03Managing token usage in long-form development sessions without losing track of current file changes

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add affaan-m/everything-claude-code strategic-compact

For use in Claude.ai and ChatGPT