How does the search agent save context window space?

Instead of loading entire raw conversation logs, the sub-agent searches, reads, and synthesizes only the relevant data, reducing context usage by 50-100x.

When is the best time to use this skill?

Use it when you are stuck on a problem, when a user references 'last time' or 'previous work', or when you need to understand the 'why' behind an existing implementation.

What is Episodic Memory in Claude Code?

It is a specialized skill that allows Claude to search through past interactions to recall specific decisions, code patterns, and solutions discussed in previous sessions.

Does this skill search my current local files?

No, this skill is specifically for conversation history. For searching your current codebase, you should use standard tools like Grep or Read.

Episodic Conversation Memory

Name: Episodic Conversation Memory
Author: jasir

byjasir

生産性とワークフロー

Recalls past conversation history and historical project decisions to provide contextual continuity and prevent reinventing solutions.

概要

Episodic Conversation Memory empowers Claude to access its historical interactions, allowing it to search through past sessions for specific architectural decisions, previously solved bugs, and established coding patterns. By utilizing a specialized search-conversations agent, it provides high-signal insights and actionable summaries from historical data without bloating the current context window. This skill is essential for maintaining momentum in long-term projects where context spans multiple sessions, ensuring that past lessons and 'gotchas' are never forgotten.

主な機能

Significant context window optimization vs. raw history loading
Dedicated search agent for context-efficient synthesis
Synthesized insights from top-ranked historical results
0 GitHub stars
Automated recognition of historical triggers and references
Intelligent search across historical conversation logs

ユースケース

Recalling specific fixes for obscure bugs encountered in previous sessions
Retrieving the reasoning behind past architectural choices during refactoring
Onboarding into unfamiliar workflows based on previous team discussions

概要

主な機能

Significant context window optimization vs. raw history loading
Dedicated search agent for context-efficient synthesis
Synthesized insights from top-ranked historical results
0 GitHub stars
Automated recognition of historical triggers and references
Intelligent search across historical conversation logs

ユースケース

Recalling specific fixes for obscure bugs encountered in previous sessions
Retrieving the reasoning behind past architectural choices during refactoring
Onboarding into unfamiliar workflows based on previous team discussions