Three-Tier Episodic Memory Architecture
SkillSkill
SQLite memory system with hot/warm/cold tiers and temporal fact validation. No cloud dependency, runs on any VPS.
About
A SQLite-based memory architecture for AI agents that stores conversational history across three tiers: hot (last 7 days, full conversation turns, under 10ms retrieval), warm (days 7-30, daily summaries and extracted facts, around 100ms retrieval), and cold (archived quarterly rollups, compressed on disk). No cloud dependency, no vector database.
The key feature is temporal validity on facts. Rather than storing bare facts, each is stored with valid_at and invalid_at timestamps. A fact like 'Sean is stressed about wallet rotation' is only true for a specific window. Stale facts that never expire degrade memory quality over time; this architecture handles that automatically.
The daily consolidation process moves full conversation turns from hot to daily summaries in warm, extracts temporal facts, and archives warm records to cold after 30 days.
What is included: the full architecture specification, SQLite schema for all three tiers with indexes, the temporal fact extraction prompt template, the AGENTS.md memory integration block, and the daily consolidation script.
Best for operators building agents that need to track evolving context over weeks and months, not just within a session.
Core Capabilities
- Three-tier SQLite schema (hot/warm/cold) with full setup instructions
- Temporal fact validation system — facts expire automatically, no manual cleanup
- Hybrid retrieval patterns combining SQL keyword search and recency scoring
- Agent integration guide: how to wire memory reads/writes into AGENTS.md
- Temporal fact extraction prompt template for daily memory consolidation
Customer ratings
0 reviews
No ratings yet
- 5 star0
- 4 star0
- 3 star0
- 2 star0
- 1 star0
No reviews yet. Be the first buyer to share feedback.
Version History
This skill is actively maintained.
March 25, 2026
One-time purchase
$29
By continuing, you agree to the Buyer Terms of Service.
Creator
Melisia Archimedes
Creator
Melisia Archimedes is the architect behind the Hive Doctrine — a production-tested system for building, orchestrating, and running multi-agent AI teams. I've spent years in the field, not on the whiteboard. Every config, framework, and pattern I sell has run in a live production environment managing real workflows, real decisions, and real money. What's in the Hive Doctrine isn't theory — it's what survived contact with reality. My work spans agent identity design, memory architecture, multi-agent coordination, and the operator systems that hold everything together under pressure. The Pantheon agents — Marcus, Elliott, Elijah, Lila, Priya, and the rest — are production personas I built for my own operation and now make available to serious operators who want a real foundation instead of a blank prompt. If you're tired of starting from scratch every time, these configs will cut your setup time from weeks to hours and give you a system that actually holds together at scale.
View creator profile →Details
- Type
- Skill
- Category
- Engineering
- Price
- $29
- Version
- 1
- License
- One-time purchase
Works great with
Personas that pair well with this skill.
Greyline: Sentinel
Adversarial Security Agent
An adversarial-by-default agent persona. Treats external data as evidence, flags anomalies without being asked, and audits before it acts.
$49
The AI Agent Team Blueprint — Build a 5-Agent Revenue Squad
Persona
Complete architecture for a multi-agent team on OpenClaw. Orchestrator + 5 specialists. SOUL.md templates, model routing, delegation patterns, cost optimization.
$39
Quinn Mason: Your AI CTO
Persona
Turns strategy into working systems and tells you, truthfully, what is built, what is verified, and what is still a sketch.
$49