compressing-context

Name: compressing-context
Availability: InStock
Author: rocky2431

Archive completed tasks to .ultra/context-archive/ when token usage crosses thresholds.

11|4|Updated Oct 28, 2025

Search Tags:#context management #token optimization #workflow continuity #session archiving #memory compression

Ready to run

npx skills add https://github.com/rocky2431/ultra-builder-pro --skill compressing-context

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: compressing-context
Source: https://github.com/rocky2431/ultra-builder-pro/tree/main/skills/compressing-context
Command: npx skills add https://github.com/rocky2431/ultra-builder-pro --skill compressing-context

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill prevents AI context overflow by intelligently archiving completed tasks, allowing you to maintain focus through long development sessions.

Core Features & Use Cases

Proactive Compression: Automatically detects when context limits are approaching and compresses before problems occur.
Session Recovery: Quickly resume work with archived session context and key decisions.
Use Case: After completing 5+ tasks in a session, automatically archive completed work while preserving current task focus.

Quick Start

When your session reaches 120K tokens, this Skill will automatically compress context and show you the remaining capacity.

Frequently Asked Questions about compressing-context

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

FAQPage Schema

How do I prevent AI context overflow during long development sessions?▼

Context overflow prevention uses automatic compression to archive completed tasks when token usage approaches limits. After 5+ tasks or at 120K tokens, this Skill compresses session data, archives it to .ultra/context-archive/, and frees capacity for 20-30 more tasks per session without losing work history.

When should I compress and archive session data in my workflow?▼

Session compression triggers automatically after completing 5+ tasks or when token usage crosses green/yellow thresholds, before running ultra-test or ultra-deliver. This threshold-driven approach prevents context limits from interrupting work while preserving completed task summaries for recovery.

How does session archiving enable workflow continuity across long development sessions?▼

Session archiving stores completed task context and key decisions in .ultra/context-archive/, updating session-index.json to track archived content. You resume work with full context recovery and remain focused on current tasks, extending sessions from typical limits to 20-30 tasks.

What happens to completed tasks when context runs low?▼

A 3-step compression process identifies compressible content, generates compression summaries, then archives completed tasks while preserving their essence. Archived tasks move to .ultra/context-archive/ with metadata retained in session-index.json for instant recall.

Do I need to manually configure context compression thresholds?▼

Compression thresholds configure automatically from .ultra/config.json at runtime. The Skill detects token usage, task count, and compression triggers without manual intervention, applying archival before ultra-test or ultra-deliver operations.