strategic-compact

Suggest manual context compaction at configurable tool-use thresholds.

Updated Dec 8, 2024
One-click install
npx skills add https://github.com/daehyunpy/my-nix-config --skill strategic-compact-daehyunpy
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: strategic-compact
Source: https://github.com/daehyunpy/my-nix-config/tree/main/home-files/cursor/skills/strategic-compact
Command: npx skills add https://github.com/daehyunpy/my-nix-config --skill strategic-compact-daehyunpy

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This Skill addresses the issue of arbitrary auto-compaction in AI chat sessions, which can disrupt workflows by losing valuable context at inconvenient moments. It promotes manual compaction at logical task intervals to preserve context more effectively.

Core Features & Use Cases

  • Manual Compaction Suggestion: Proactively suggests when to manually compact the AI's context.
  • Phase-Based Context Preservation: Helps maintain relevant information through different stages of a task (e.g., exploration vs. execution).
  • Configurable Thresholds: Allows customization of when suggestions are made based on tool usage.
  • Use Case: After extensive research and planning for a coding task, you can use this skill's suggestion to compact the research context, keeping the implementation plan fresh for the next phase.

Quick Start

Ensure the hook is set up in your ~/.claude/settings.json to receive suggestions for manual compaction.

Frequently Asked Questions about strategic-compact

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

FAQPage Schema
How do I prevent auto-compaction from losing context during AI chat sessions?

Manual context compaction prevents auto-compaction from losing context during AI chat sessions by prompting you to compact at logical task intervals. This ensures important information is retained during critical task transitions.

When should I compact context to preserve task phases in my workflow?

You should compact context to preserve task phases when transitioning between stages like exploration and execution. The skill suggests manual compaction at logical intervals, keeping implementation plans fresh after research.

How do I configure context compaction thresholds based on tool usage?

You can configure context compaction thresholds based on tool usage by setting up a hook in your settings file. The skill tracks tool calls and prompts you to manually compact when a configurable threshold is reached.

Why does arbitrary auto-compaction disrupt coding workflows?

Arbitrary auto-compaction disrupts coding workflows by losing valuable context at inconvenient moments. Manual compaction at logical task intervals preserves relevant information more effectively across different task stages.

Can I control when the AI assistant suggests manual context compaction?

Yes, you can control when the AI assistant suggests manual context compaction through configurable thresholds. The skill tracks tool calls and prompts you at periodic intervals after a threshold is reached.

Do I need a specific environment setup to receive manual compaction suggestions?

You need to set up a hook in your settings file to receive suggestions for manual compaction. This allows the skill to track tool calls and prompt you when the configured threshold is reached.