remembering-conversations

What problem does it solve?

This skill provides Claude with perfect recall of all past conversations and projects, preventing the need to reinvent solutions or repeat past mistakes due to lack of memory between sessions. It significantly reduces context window usage by synthesizing findings from historical data.

Core Features & Use Cases

• Mandatory Search Agent: Dispatch a specialized subagent to semantically search episodic memory for relevant past conversations.
• Context-Efficient Recall: The agent synthesizes findings from top results, saving 50-100x context compared to loading raw conversations.
• Proactive Search: Use before implementing features, making architectural decisions, debugging, or starting any non-trivial task.
• Semantic & Text Search: Search by single or multi-concept queries, with options for vector, text, or combined modes.
• Date Filtering: Filter search results by date ranges for precise historical context.
• Use Case: Before starting a new feature, dispatch the search agent to find past discussions, code examples, or "gotchas" related to similar implementations, ensuring you build on prior knowledge.

Quick Start

Dispatching search agent to find past discussions about React authentication. Task tool: description: "Search past conversations for React authentication patterns" prompt: "Search for React authentication patterns. Focus on decisions, gotchas, and code examples." subagent_type: "search-conversations"