compaction-skill

What problem does it solve?

It prevents AI memory from growing beyond practical token limits by converting verbose conversation history into a concise, structured summary that retains the information needed to continue effectively.

Core Features & Use Cases

• Token-saving memory compaction: Triggers when token thresholds are near and compresses prior messages into a summary.
• Structured summaries for continuity: Produces a consistent 5-section output (Task Overview, Current State, Important Discoveries, Next Steps, Context to Preserve) designed for seamless handoff into subsequent conversation turns.
• Decision and context preservation: Captures what was decided, why it was decided, what worked or failed, and specific critical details (values, IDs, preferences, constraints) to avoid repeating effort.

Quick Start

Ask your assistant to run compaction and replace the current conversation memory with a structured summary using the five required sections when the token budget is approaching its limit.