How does the skill suggest a compact?

The skill uses a script that monitors your tool calls. Once you reach a threshold (default is 50 calls), it provides a reminder that it might be a good time to run the /compact command.

Why use manual compression instead of automatic?

Automatic compression often triggers at arbitrary points, which can cause the AI to lose track of variables or reasoning mid-task. Manual compression at logical boundaries preserves focus.

Can I customize the alert frequency?

Yes, you can set the COMPACT_THRESHOLD environment variable to define the number of tool calls that occur before the first suggestion appears.

What information is kept after a strategic compact?

Core project state remains intact, including CLAUDE.md instructions, your TodoWrite task list, memory files (~/.claude/memory/), and your Git status.

Strategic Context Compactor

Name: Strategic Context Compactor
Author: saar120

bysaar120

0•

生产力与工作流

Optimizes Claude Code sessions by suggesting manual context compression at logical task boundaries to preserve critical information.

The Strategic Context Compactor skill improves Claude Code session performance by replacing arbitrary automatic compression with intelligent, milestone-based manual prompts. By monitoring tool usage and session density, this skill alerts developers at strategic transition points—such as moving from research to implementation or finishing a debugging sprint—ensuring that essential reasoning and project state are preserved while maintaining high response speeds in long-running sessions exceeding 200K tokens.

主要功能

010 GitHub stars

02Phase-specific compression guides for research, planning, and execution

03Intelligent tool-call tracking to monitor session context density

04Configurable threshold alerts for proactive context management

05Seamless integration with PreToolUse hooks for real-time monitoring

06Advanced token optimization patterns to reduce baseline overhead

使用场景

01Managing performance and coherence in long-running sessions with heavy token usage

02Transitioning between multi-stage tasks like research, planning, and implementation

03Cleaning up messy debugging traces before starting a new feature to prevent context pollution

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add saar120/skillgate-registry strategic-compact

For use in Claude.ai and ChatGPT

主要功能

010 GitHub stars

02Phase-specific compression guides for research, planning, and execution

03Intelligent tool-call tracking to monitor session context density

04Configurable threshold alerts for proactive context management

05Seamless integration with PreToolUse hooks for real-time monitoring

06Advanced token optimization patterns to reduce baseline overhead

使用场景

01Managing performance and coherence in long-running sessions with heavy token usage

02Transitioning between multi-stage tasks like research, planning, and implementation

03Cleaning up messy debugging traces before starting a new feature to prevent context pollution

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add saar120/skillgate-registry strategic-compact

For use in Claude.ai and ChatGPT