unclick-memory

What problem does it solve?

It ensures an AI agent always has up-to-date, persistent user context so answers and decisions reflect the user’s real preferences, facts, and ongoing work instead of forgetting between sessions.

Core Features & Use Cases

• Session bootstrapping with authoritative memory: Automatically loads identity, standing rules, preferences, recent sessions, and active facts at the very start of each interaction.
• Context-aware retrieval on demand: Searches stored memory when the user references past items like preferences, “last time,” identity, facts about them, or related projects/people/tools.
• Proactive persistence of new information: Saves new facts and standing rules as the user provides them, and records session summaries and checkpoints so future sessions pick up where the last left off.
• Common triggers without user prompting: Reacts immediately to phrases like “remember,” “recall,” “who am I,” and context-related mentions to minimize missing personalization.

Quick Start

Trigger the unclick-memory skill at the very start of every session, then call load_memory before the first response and follow the LOAD → SEARCH → SAVE → END protocol throughout the conversation.