oneiros

Manage persistent agent memory and cognitive context with an event-based SQLite system.

3|Updated Jan 15, 2026
One-click install
npx skills add https://github.com/esmevane/oneiros --skill oneiros
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: oneiros
Source: https://github.com/esmevane/oneiros/tree/main/crates/oneiros-skill/skill
Command: npx skills add https://github.com/esmevane/oneiros --skill oneiros

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This Skill provides persistent memory and cognitive context for AI agents, allowing their "brains" to survive across sessions and context compactions.

Core Features & Use Cases

  • Persistent Memory: Agents retain knowledge and experiences beyond a single session.
  • Cognitive Loops: Manages agent lifecycle through dreaming, introspection, and reflection.
  • Multi-Agent Coordination: Facilitates collaboration and context sharing among agents.
  • Use Case: Ensure your AI assistant remembers your preferences and past interactions, even after a system restart or model update, by using Oneiros to manage its long-term cognitive state.

Quick Start

Use the oneiros skill to dream the governor agent's full cognitive context.

Frequently Asked Questions about oneiros

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

FAQPage Schema
How do I maintain persistent memory for an AI agent across multiple sessions?

You can maintain persistent agent memory across sessions by using an event-based SQLite system that stores cognitive context, allowing AI agents to retain knowledge and experiences continuously even after system restarts or context compactions.

What is agent session continuity and how does event-based memory work?

Agent session continuity is the ability for an AI to preserve its cognitive state across interactions. It works by using an event-based SQLite system to log experiences, enabling layered memory, introspective reflection, and emergent personas over time.

Can I use this memory system for multi-agent coordination and context sharing?

Yes, you can use this persistent memory system for multi-agent coordination. It facilitates collaboration and context sharing among multiple agents by managing their cognitive states within a unified event-based SQLite architecture.

Does this agent memory system work with different AI models?

Yes, the agent memory system supports model-agnostic operation. This means you can maintain persistent cognitive context and long-running continuous brains for agents regardless of the specific AI model you are using.

How do I start an agent's cognitive loop for dreaming and introspection?

You can start the cognitive loop by triggering the skill to dream the governor agent's full cognitive context. This manages the agent lifecycle through continuous cycles of dreaming, introspection, and reflection.

What are the limitations of using SQLite for cognitive persistence in AI agents?

The metadata does not specify explicit limitations of SQLite for cognitive persistence. However, the system is designed to handle event-based memory and context layering to prevent data loss during context compaction.