memory-consolidation

🧠 The memory that dreams — cross-session memory for DeepSeek Harness. Offline & private, auto-consolidates in its sleep (autoDream), visualized in a memory panel.

Agent memory for LLM agents: 7 neuroscience-inspired layers (working, episodic, semantic, procedural) with FSRS spaced repetition and memory consolidation. Zero-dependency TypeScript library, Model Context Protocol (MCP) server, Vercel AI SDK middleware. Wins 9/9 answer-quality comparisons on LongMemEval-500 (3 judges, Bonferroni).

Machine Dream - Continuous machine cognition research platform with GRASP loop, dreaming consolidation, and persistent memory. Pure LLM Sudoku solver that learns from experience and consolidates knowledge through AI "sleep cycles."

Plugins as organs for DeepSeek Harness: nerve impulses route each command, a heartbeat circulates state, reflex arcs fire with zero model calls, sleep-time memory, closed-loop self-healing. Tool-schema gating cuts prompt tokens 84.7% (reproducible).

Brain-inspired cognitive architecture implementing basal ganglia RL, hippocampal memory consolidation, and prefrontal meta-cognition. Multi-agent system with dynamic attention control, procedural learning, and theory of mind - genuine cognitive continuity beyond context windows.

dreamweave - a nightly memory engine for OpenClaw-based agents (incl. Microsoft Scout). Your agent dreams: it weaves a brain-faithful graph+vector memory and consolidates it like sleep - forgetting curve, merge, temporal weave - across three tiers. Independent community project; not affiliated with or endorsed by Microsoft.

The continuity layer for everything you do with AI. A self-hosted, open source server any tool plugs into over MCP or REST: hybrid vector + lexical recall, an auto-built knowledge graph, sleep-style consolidation, procedural and persona tiers, OAuth 2.0, multi-tenant. SQLite or Postgres. MIT.

Always-on AI agent with persistent long-term memory, sleep-time consolidation, and dreams — runs 24/7 on a small local LLM (Ollama, 2B params, zero dependencies)

Selective, imperfect, reconstructive memory for LLMs

Deep learning implementation of the Benna-Fusi complex synapse model.

OpenDream — open-source memory consolidation for AI agents. Reads past any LLMs sessions, dreams across them to extract patterns, and writes consolidated memory into AGENTS.md so the next session starts smarter. Local SQLite, BYO LLM, no SaaS.

The memory database for AI agents - graph memory, hybrid search, tamper-evident thought/edge journal.

Dreaming plugin for OpenClaw memory-lancedb — Light/REM/Deep sleep phases with narrative generation and memory promotion.

Self-audit memory engine for Hermes operators. Analyzes dialogues, extracts durable corrections, snapshots and updates USER.md/MEMORY.md safely.

Totemheart is a deterministic control kernel for persistent cognition and relational behavior, not a sentiment classifier and not a prompt-engineering trick.

Memory consolidation skills for Claude Code — read-only dream pass, checkbox/validator gate, full-auto mode, snapshot + one-command rollback. No rm, ever.

Nightly deterministic memory consolidation ("dreaming") skill for Hermes Agent

adaptive memory layer for agentic AI

Testing whether selective forgetting improves LLM agent coherence over long conversations.

🧠 AI Agent Memory Layer — 双层记忆架构，让LLM Agent拥有持久记忆。支持语义搜索、记忆巩固、防污染检测。每天在生产环境运行。