azoth-memory-capture

Captures session lessons as structured episodes with append-only governance and cross-linkages.

Updated Apr 8, 2026
One-click install
npx skills add https://github.com/yiwei79/azoth --skill azoth-memory-capture
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: azoth-memory-capture
Source: https://github.com/yiwei79/azoth/tree/main/.agents/skills/azoth-memory-capture
Command: npx skills add https://github.com/yiwei79/azoth --skill azoth-memory-capture

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Capture durable lessons from each Azoth session and preserve them as structured episodes to prevent knowledge loss.

Core Features & Use Cases

  • Write interface to Azoth's 3-layer memory system, capturing episodes (M3), surfacing patterns for promotion to M2, and contributing to procedural knowledge (M1).
  • Enforce episode governance with append-only editing, explicit linking between episodes, and tagging for reliable recall.
  • Support automated surfacing and promotion workflows, including auto-classification and promotion proposals when patterns are reinforced across sessions.

Quick Start

Capture a new episode at session end to archive lessons and trigger auto-classification

Frequently Asked Questions about azoth-memory-capture

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

FAQPage Schema
How do I capture session lessons to prevent knowledge loss in AI memory?

You can capture session lessons by archiving them as structured episodes using an append-only schema with timestamps, types, summaries, and tags to prevent knowledge loss. This automatically surfaces recurring patterns for higher memory layer promotion.

What is the best way to enforce governance rules on episodic memory records?

The best way to enforce governance on episodic memory is using an append-only editing model that creates explicit cross-episode linkages for supersedes and related relationships. This ensures reliable recall and prevents unauthorized historical modifications.

How does auto-classification promote recurring patterns across sessions?

Auto-classification promotes recurring patterns by evaluating captured episodes across sessions and proposing elevation to higher memory layers when behaviors are reinforced. This automated workflow transforms isolated session lessons into durable procedural knowledge.

When should I archive decisions using an episode schema?

You should archive decisions using an episode schema during session closeout or immediately after significant decisions are made. This timing captures the exact context, lessons, and tags needed to prevent knowledge loss before the session ends.

Can I link related episodes together to preserve context across multiple sessions?

Yes, you can link related episodes together using explicit cross-episode relationships for supersedes and related connections. This structured linkage preserves context across multiple sessions and ensures reliable recall of evolving knowledge.