Remembering Conversations

This skill prevents Claude from wasting time reinventing solutions you've already discussed. When you ask "how should I implement X" or reference past conversations, it dispatches a specialized search agent that scans conversation history and synthesizes relevant findings in 200-1000 words. The agent uses episodic memory tools to find previous architectural decisions, gotchas you've hit, and code patterns that worked. Much smarter than loading raw conversation transcripts into context. Saves massive amounts of back-and-forth when you're building on previous work or facing problems you've solved before. Essential for any ongoing project where context matters.

skills.sh installs: 9.2K
GitHub stars: 499
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add obra/episodic-memory --skill remembering-conversations --agent claude-code

Installs into .claude/skills of the current project.

Files

Remembering Conversations

Core principle: Search before reinventing. Searching costs nothing; reinventing or repeating mistakes costs everything.

Mandatory: Search Historical Memory

YOU MUST search historical memory for any historical search.

Announce: "Searching past conversations for [topic]."

Claude Code

Use the Task tool with subagent_type: "search-conversations":

Task tool:
  description: "Search past conversations for [topic]"
  prompt: "Search for [specific query or topic]. Focus on [what you're looking for - e.g., decisions, patterns, gotchas, code examples]."
  subagent_type: "search-conversations"