Structured memory for agents: weighted retrieval and replayable evidence paths
MCP server that extends traditional knowledge graphs with structural tension charts enabling coaia-spiral persistent memory for Claude through a local knowledge graph - fork focused on local development
MemU is an agentic memory framework for LLM and AI agent backends: it ingests multi-modal data, extracts and organizes it into structured memory and supports both RAG and LLM-based retrieval.
Modular Python AI system featuring FastAPI, Ollama, persistent memory (SQLite/ChromaDB), RAG, n8n automation, and 187 passing tests.
A Codex skill for keeping long AI work on track with summaries, structured memory, and timely clarification.
Xytha Nous 1.0: a runnable, source-traceable domain agent with chart tools, memory and recovery. Plus Memory Atlas, a six-layer memory playground.
MCP server for project memory — AI agents record decisions, pitfalls, and specs as they work, and recall them next session with code context. Built for agents, not humans.
Experimental local‑first AI runtime with multi‑clock reasoning and structured memory consolidation. Designed for extensible LLM workflows and reusable reasoning patterns.
Free structural-memory pipeline for AI agents. SO Extractor Free turns conversation logs into MemoryUnits; SO Memory Engine builds Context Packs exposing recurrence, tensions, connections, and evidence identity before an LLM responds.
⚛️ Atomic Blueprint (ab) - The App-Dev Agent Memory Kernel. Give your AI agents a memory that actually remembers.
Low-latency runtime for structured conversational processes - process DAG, sole-writer state, observations as LTM, portable model adapter.
RefNet is a 2M-parameter edge-aware transformer for structured introspection and reflective evaluation within Structured Reflective Cognitive Architecture (SRCA/SRAI) systems. It predicts cognitive metrics (valence, self-model drift, thought quality) and recommends introspective actions (consolidate, recall, reframe, evaluate_alignment)
MemCube — Structured Memory Management Skill for OpenClaw (inspired by MemOS MemTensor)
Practical methodology for engineering structured long-term memory for AI agents using MemPalace.
Local-first persistent memory for Codex with hook-based recall, decay-aware retrieval, structured memory, topic regrouping, and an Obsidian mirror.
SINT-UA v2.1: Agent memory system for OpenCode with hash-chain provenance, temporal decay, drift detection, and provenance archaeology
To associate your repository with the structured-memory topic, visit your repo's landing page and select "manage topics."