mnemonic-core

Manage persistent filesystem-based memory stores compliant with Memory Interchange Format Level 3.

20|4|Updated Jan 24, 2026

Search Tags:#knowledge management #recall #ai agent #memory #persistence #capture

Ready to run

npx skills add https://github.com/zircote/mnemonic --skill mnemonic-core

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: mnemonic-core
Source: https://github.com/zircote/mnemonic/tree/main/skills/mnemonic-core
Command: npx skills add https://github.com/zircote/mnemonic --skill mnemonic-core

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill provides a robust, filesystem-based memory system for AI agents, enabling persistent storage and retrieval of information across conversations and sessions.

Core Features & Use Cases

Memory Capture: Silently captures user-stated decisions, learnings, patterns, and blockers into structured markdown files.
Memory Recall: Enables searching and retrieving memories using keywords and context.
Use Case: When discussing a complex technical problem, the Skill can automatically save the root cause and its solution as a "learning" memory, which can be recalled later if a similar issue arises.

Quick Start

Use the mnemonic-core skill to capture the current decision about using JWT for API authentication.

Frequently Asked Questions about mnemonic-core

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

FAQPage Schema

How do I add persistent memory to an AI agent across sessions?▼

A persistent memory store captures user decisions, learnings, and patterns into structured markdown files, enabling AI agents to recall information across conversations and sessions.

How does bi-temporal tracking work for AI memory capture?▼

Bi-temporal tracking records both when an event actually occurred and when the memory was stored in the filesystem. This allows precise chronological recall of semantic, episodic, and procedural memories.

Can I use vector search to retrieve AI agent memories?▼

Yes, optional vector search integration is supported for retrieving memories. This supplements standard keyword and context searching to find relevant semantic, episodic, or procedural data.

What is the best way to capture technical decisions and learnings for an AI agent?▼

Use a Memory Interchange Format compliant system to silently capture technical decisions and learnings. This automatically saves root causes and solutions as structured memories for later recall.

Do I need external dependencies to maintain a knowledge management system for my AI agent?▼

No external dependencies are required. The memory system operates using standard Unix tools and Claude's native capabilities to manage filesystem-based memory operations and custom ontologies.