An open, vendor-neutral AI memory format and protocol
An open, vendor-neutral AI memory format and protocol
Portable Markdown templates to own your AI memory across ChatGPT, Gemini, Claude and other AI tools.
Persistent, portable, shared memory for Massive Intelligence (IM) agents: pluggable stores, a deterministic identity seed stable across model swaps, masking and residency, and Ed25519-signed, SHA-256-chained tamper-evident access logs. Zero dependencies, node-free core, TypeScript-first.
Stop explaining yourself to every AI. User-authored memory with a private link and optional public profile. Works with ChatGPT, Claude, Gemini. https://aimemlink.com
Portable memory stack for Codex with Markdown workspace logs, MCP semantic recall, and a reusable skill workflow.
Tiny portable memory format for agents: JSONL plus markdown plus MCP examples
本地经验复盘与复用。核心特色：跨设备/跨客户端的经验整合 + 整档案的备份与恢复；经验只存本机、不上传，明文可编辑、可审计、可按需合并。Local-first experience layer for AI agents: merge your experience across devices and clients, back it up, never upload. 关键词：经验复盘、经验沉淀、长期记忆、个性化、跨 Agent 迁移。
Animated Android AI assistant with voice and portable encrypted memory. Contributors wanted for Google Play, iOS and local LLM support. Community alpha; MIT code, reserved artwork.
Encrypted and signed portable personal-memory capsule specification
Owned, portable working memory: a plain-files brain template + a tiny MCP server so any model can read it. Files, not a login. From br8n, the AI delivery practice of Branded Mayhem Collective.
Portable AI memory that roams with you across models and harnesses — Claude Code, Codex, Cursor, ChatGPT. Plain markdown you own, wired over MCP with one command.
To associate your repository with the portable-memory topic, visit your repo's landing page and select "manage topics."