M

memory

Persistent semantic memory system for OpenCode sessions - MCP server with hybrid FTS/vector search

Gives a coding agent the memories that bear on the work in hand, from files on the machine, in Claude Code and in Grok.

A generic, concurrent, portable and flexible Object Pool for .NET: https://object-pool-docs.pommalabs.xyz/

Lintorn t'aide à ne pas te perdre dans ton projet : il vérifie que ta doc et la mémoire de ton IA disent encore la vérité sur ton code. Tu lui donnes tes règles, il les fait respecter à chaque git push, et il installe les bons outils d'audit selon tes langages. Aucunement intrusif : il ne lance rien sans ton accord. À essayer : pipx install lintorn. Projet jeune, créé seul avec un assistant IA : pas encore stable à 100 %.

Level-anchored memory-pressure tuning for NixOS: zram/zswap sizing, PSI-armed systemd-oomd, and vm sysctls from one declared RAM level

Pioche fabrique vos flashcards : une question d'un côté, sa réponse de l'autre. Vous les saisissez à la chaîne, vous les rangez par matière et par cours, et vous les imprimez au format d'une carte à jouer, prêtes à découper

Recall MCP — long-term memory for AI agents. Cross-session, zero-dependency, MCP-native. SQLite storage, BM25 search, and memory decay. Works with Claude, Cursor, and any MCP client.

CLI tools for working with .aleth encrypted memory files — inspect, verify, and search agent memory without decrypting the whole chain. Published on npm as alethech-tools.

A CLI and Rust library for obtaining your firmware's AGESA version on Linux.

SynapCache is a distributed, zero-memory-loss caching system for all LLM outputs, designed to be LLM-agnostic, high-performance, and scalable. It combines TurboQuant-style compression, semantic search, edge caching, and predictive intelligence to store and retrieve outputs efficiently, ensuring instant recall and minimal resource usage. With developer-friendly SDKs, plugin support, and multi-cloud deployment, SynapCache provides a universal neural memory layer for modern AI workflows. https://roxanneardary.com/synapcache/

GRU is an Aonz.ai chat and PDF retrieval Chatbot with a React/Vite frontend and a FastAPI backend. It routes Aonz.ai questions through a Chroma knowledge base and sends unrelated questions directly to an OpenAI-compatible LiteLLM proxy. Requests are traced with Langfuse and responses can be cached in Valkey using exact and semantic matching.

Rain World early warning system for out of memory related incidents.

Idea2Reality is an AGPL-3.0+ open-source industrial intelligence platform that transforms ideas into manufacturable hardware and software systems, optimizing across manufacturing, logistics, geography, labor, and cost. It integrates AI-driven design, real-world feasibility analysis, 3D system visualization, workforce and apprenticeship mapping, and continuous memory-based learning to generate end-to-end production and execution plans. https://roxanneardary.com/idea2reality/

EchoCore Memory is an open source AGPL 3.0+ modular memory orchestration system for Linux that transforms RAM management from passive allocation into intelligent, policy-driven control. By combining memory scheduling, adaptive swap strategies, per-application memory profiles, workload modes, and optional AI-based prediction, EchoCore Memory helps systems remain responsive, stable, and efficient under changing workloads. https://roxanneardary.com/echocore-memory/

Composable Agent primitives for Kit applications

Primitives and generic deriving for OpenGL/Khronos Standard Uniform Block Layout

claude-cortex is a file-based memory and behavior layer for Claude Code, the VS Code extension, and Xcode. No cloud sync, no lock-in — just structured Markdown that travels with your project and keeps Claude context-aware across every session.

🧠 Open-source persistent memory engine for AI agents and LLMs with semantic search, automatic deduplication, and intelligent context retrieval.

A hybrid cache implementation inspired from 'https://github.com/dotnetcore/EasyCaching' with Go