memory-engine

Manage a persistent SQLite memory store with HNSW vector indexing.

18|7|Updated Mar 11, 2026

Search Tags:#context management #ai agent #memory #sqlite #persistence #vector search

Ready to run

npx skills add https://github.com/thecloudtips/founder-os --skill memory-engine

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: memory-engine
Source: https://github.com/thecloudtips/founder-os/tree/main/plugin/skills/infrastructure/memory
Command: npx skills add https://github.com/thecloudtips/founder-os --skill memory-engine

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill provides a robust, persistent memory store that allows AI agents to learn from user interactions, maintain context across sessions, and personalize their behavior over time.

Core Features & Use Cases

Cross-Plugin Memory: A unified SQLite database accessible by all Founder OS plugins.
Structured Storage: Stores user preferences, detected patterns, business facts, contact details, and workflow knowledge.
Semantic Search: Utilizes HNSW vector indexing for efficient similarity searches on memory content.
Use Case: When a user consistently asks to archive newsletters, the memory engine stores this as a preference memory. Future interactions can then automatically archive newsletters without explicit instruction.

Quick Start

Initialize the memory store if it doesn't exist by running the provided bash command.

Frequently Asked Questions about memory-engine

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 multiple sessions?▼

Persistent memory for an AI agent is created by storing user preferences and behavioral patterns in a SQLite database with HNSW vector indexing. This enables the agent to maintain context and personalize interactions across different sessions.

What is semantic search for AI context management and when is it needed?▼

Semantic search for AI context management uses HNSW vector indexing to find similar content in stored memory. It is needed when an AI agent must retrieve previously learned user preferences, business facts, or workflow knowledge based on meaning rather than exact keywords.

How do I store and retrieve user preferences in an AI memory store?▼

You store user preferences by saving them as structured entries in a SQLite memory database, then retrieve them using semantic similarity search or exact key lookups. The system detects behavioral patterns and stores them for future automated retrieval.

Can I use SQLite for vector search and context management in AI agents?▼

Yes, SQLite can be used for AI agent context management by combining it with HNSW vector indexing. This setup supports efficient similarity searches on memory content, alongside structured storage for business facts and contact information.

What is the best way to query company-specific facts from an agent memory database?▼

The best way to query company-specific facts is using a persistent memory store that supports company-specific memory retrieval alongside semantic search. This allows an AI agent to filter and recall business facts relevant to a specific organization.

Does the memory engine work without external dependencies?▼

The memory engine operates with no external dependencies, utilizing a built-in SQLite database for structured storage and HNSW vector indexing for semantic search. It initializes directly via a bash command when the memory store does not yet exist.