context-compression

Compress long-running sessions using anchored summarization and observation masking.

11|1|Updated Jul 5, 2010

Search Tags:#cache #compression #compaction #masking

Ready to run

npx skills add https://github.com/georgeguimaraes/dotfiles --skill context-compression-georgeguimaraes

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: context-compression
Source: https://github.com/georgeguimaraes/dotfiles/tree/main/dot_claude/skills/context-engineering/context-compression
Command: npx skills add https://github.com/georgeguimaraes/dotfiles --skill context-compression-georgeguimaraes

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This skill provides strategies for compressing long-running sessions while preserving essential information: anchored summarization, observation masking, and cache-aware techniques.

Core Concepts

Compaction: merge and summarize to stay within limits
Observation Masking: replace verbose tool outputs with compact references
KV-Cache Optimization: maximize cache hits for efficiency
Context Partitioning: isolate tasks to reduce cross-talk

Quick Start

Use the compression evaluator to assess token savings vs signal loss on a sample history.