rekal-save

Capture session knowledge and deduplicate it via rekal MCP tools.

52|4|Updated Apr 11, 2026
One-click install
npx skills add https://github.com/janbjorge/rekal --skill rekal-save
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: rekal-save
Source: https://github.com/janbjorge/rekal/tree/main/skills/rekal-save
Command: npx skills add https://github.com/janbjorge/rekal --skill rekal-save

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Sessions often generate valuable preferences, decisions, and procedures that are lost after the conversation ends. This skill captures that durable knowledge, removes duplicates, and stores it for future sessions so users never repeat themselves.

Core Features & Use Cases

  • Candidate extraction: Scans the conversation for knowledge worth remembering.
  • Deduplication & superseding: Prevents multiple memories on the same topic, updating outdated or contradictory information.
  • Project scoping & conflict detection: Associates memories with a project and flags inconsistencies for review.

Quick Start

Run the rekal-save skill at the end of a chat to capture and store key insights.

Frequently Asked Questions about rekal-save

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I save session insights and preferences for future use?

Session insights and preferences are saved by scanning the conversation for durable knowledge, removing duplicates, and storing it for future retrieval so you never repeat yourself.

What's the best way to deduplicate knowledge and prevent conflicting memories?

Deduplicating knowledge is handled by identifying overlapping topics, superseding outdated information, and flagging contradictory inconsistencies for review before storing.

Can I associate stored memories with a specific project scope?

Project scoping is supported by associating captured memories with a specific project and detecting conflicts to ensure correct contextual retrieval.

When should I capture and store conversation knowledge during a chat?

Capturing conversation knowledge is ideal at end-of-session or mid-session moments when important preferences, decisions, or procedures arise.

Do I need specific memory tools to store and deduplicate session knowledge?

Storing session knowledge requires access to memory MCP tools including memory_search, memory_store, memory_supersede, memory_conflicts, and memory_set_project.

Why does my stored knowledge contain duplicate or contradictory preferences?

Duplicate or contradictory preferences occur when knowledge is stored without deduplication, which this process prevents by superseding outdated information and flagging conflicts.