vector-memory

🧠 Production-grade memory sidecar for AI agents — gbrain + Hindsight + 3-tier recall. Agent-agnostic, battle-tested. | 生产级外挂记忆系统，兼容Hermes/Claude/Cursor等任意AI智能体

Architectural enhancement suite for Hermes Agent: vector memory, adaptive soul, and workflow classification.

Self-hosted personal AI assistant on Telegram. Powered by OpenCode + Claude Sonnet (free big-pickle). OpenClaw skills compatible. SQLite + MCP live memory with optional Ollama vector search.

AI agent memory using Modern Hopfield Networks — no LLM calls, no database, one matrix multiply. MCP server for Cursor, Claude Code, and other AI coding agents.

Open-source personal AI agent that runs on your own laptop, CPU-only. Self-improving and self-learning. Every outcome it produces is verified and logged — the evidence ledger is public-recordable. Windows-native.

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

An Eggdrop IRC bot script that pipes LLM intelligence into classic chat

Cortex — self-evolving memory and agent fleet for Claude Code

A sturdy, multimodal AI desktop companion for Windows. Features long-term memory, multi-profile personas, voice interactions, and local tools. Powered by Gemini 2.5. 🐴✨

Agent memory runtime: short/long-term context, vector persistence, compression, and personalization primitives.

An Fully local, privacy-first AI chat application. It features persistent vector memory, website crawling & Q&A, and fully customizable chatbot persona.

EVM — Evolution Vector Memory: deterministic interaction-state continuity and bounded identity evolution framework for AI systems.

Lilith CLI - terminal-first AI workspace with a Textual IDE, durable missions, orchestration, and vector memory.

🧠 ViBo — save 50-150× tokens: persistent memory for AI agents (L1/L2/L3) + web search compression (96.2%). One skill, two products. $5/mo.

AIOS is an operating system for language models and human users that runs like a text-based adventure game for intelligent agents. It is feature-rich with a working memory pipeline.

A tool-calling AI agent with a hand-written loop: sandboxed code execution, scoped file I/O, web search, long-term vector memory, and failure guardrails (circuit breaker, iteration limits). Keyless mock LLM or Groq. Python/FastAPI.

Telegram-бот с долговременной памятью в Pinecone: чат, RAG и фильтр дубликатов по косинусному сходству

Session memory & context enhancer for chat.z.ai — MCP server (12 tools), local vector memory, token saver, smart skills, GitHub private-repo cloud DB. Built with GLM 5.3 Flash.

A self-evolving, tri-brain autonomous AI agent featuring local subconscious memory, generative SVG art, and conscious reasoning.

Plug-and-play Qdrant vector memory plugin for Hermes Agent — 8 tools, 10 modules, 58 tests. Semantic long-term memory via local Qdrant + any OpenAI-compatible embedding API. Self-healing. One-minute install.