LLMemory – Backup ChatGPT, Claude & Gemini 作者: LLMemory
Export, back up and search your AI chats — ChatGPT, Claude, Gemini and more — into your own private, searchable memory on Drive.
擴充套件後設資料
畫面擷圖
關於此擴充套件
Back up, export, and search every AI conversation you have into storage you control. LLMemory saves your chats to your own Google Drive (or a private GitHub repo) as plain JSON. Nothing touches our servers. Your chats, your storage, your keys.
WHY LLMEMORY INSTEAD OF A ONE-OFF EXPORT
Most exporters dump a single file and forget it. LLMemory builds a living, searchable archive of your entire AI chat history that grows with you:
Import once, own forever — one click backs up years of conversations into your vault. No manual export, no copy-paste.
Live capture — every new conversation is saved automatically as you chat. Turn it on and forget it.
Full-text search — find anything across every chat in seconds, titles and messages alike.
Your storage, your rules — your vault lives in your Google Drive or a private GitHub repo. No vendor lock-in, no account database — just JSON files in a folder you own. Switch or download anytime.
Zero-knowledge by design — the extension never sends your chat content to our servers. The OAuth token exchange is the only external call, and it's stateless (stores nothing).
Cross-sync Drive and GitHub — mirror your vault for redundancy; GitHub keeps full version history of every change.
WORKS EVERYWHERE YOU USE AI
Browser extension — capture your conversations live, or backfill your entire history in one click.
CLI (npm i -g llmemory-cli) — back up Codex and Claude Code sessions from your machine, push to Drive, and search from the terminal.
Web vault (app.llmemory.xyz) — read and search your archive from any browser; cross-sync Drive and GitHub for redundant backups.
MCP server — coding agents like Claude Code can read your prior conversations live; one-command setup with llmemory connect-claude.
PRIVACY-FIRST
All chat content stays on your machine and in your own cloud storage.
OAuth tokens are stored locally, never transmitted to third parties.
The token-exchange Worker is stateless — it sees codes, returns tokens, stores nothing.
No analytics, no trackers, no account database.
WHAT IT CAPTURES
AI chat assistants — ChatGPT and Claude.ai are saved with full message trees, timestamps, model info, and conversation branches. Other major web-based AI assistants are backed up automatically as you chat.
Developer CLIs — Claude Code sessions (from ~/.claude/projects) and Codex sessions (from ~/.codex), backed up straight from your machine.
WHY LLMEMORY INSTEAD OF A ONE-OFF EXPORT
Most exporters dump a single file and forget it. LLMemory builds a living, searchable archive of your entire AI chat history that grows with you:
Import once, own forever — one click backs up years of conversations into your vault. No manual export, no copy-paste.
Live capture — every new conversation is saved automatically as you chat. Turn it on and forget it.
Full-text search — find anything across every chat in seconds, titles and messages alike.
Your storage, your rules — your vault lives in your Google Drive or a private GitHub repo. No vendor lock-in, no account database — just JSON files in a folder you own. Switch or download anytime.
Zero-knowledge by design — the extension never sends your chat content to our servers. The OAuth token exchange is the only external call, and it's stateless (stores nothing).
Cross-sync Drive and GitHub — mirror your vault for redundancy; GitHub keeps full version history of every change.
WORKS EVERYWHERE YOU USE AI
Browser extension — capture your conversations live, or backfill your entire history in one click.
CLI (npm i -g llmemory-cli) — back up Codex and Claude Code sessions from your machine, push to Drive, and search from the terminal.
Web vault (app.llmemory.xyz) — read and search your archive from any browser; cross-sync Drive and GitHub for redundant backups.
MCP server — coding agents like Claude Code can read your prior conversations live; one-command setup with llmemory connect-claude.
PRIVACY-FIRST
All chat content stays on your machine and in your own cloud storage.
OAuth tokens are stored locally, never transmitted to third parties.
The token-exchange Worker is stateless — it sees codes, returns tokens, stores nothing.
No analytics, no trackers, no account database.
WHAT IT CAPTURES
AI chat assistants — ChatGPT and Claude.ai are saved with full message trees, timestamps, model info, and conversation branches. Other major web-based AI assistants are backed up automatically as you chat.
Developer CLIs — Claude Code sessions (from ~/.claude/projects) and Codex sessions (from ~/.codex), backed up straight from your machine.
由 1 位評論者給出 0 分
權限與資料
必要權限:
- 存取瀏覽器分頁
- 存取您在 chatgpt.com 的資料
- 存取您在 chat.openai.com 的資料
- 存取您在 claude.ai 的資料
- 存取您在 gemini.google.com 的資料
- 存取您在 www.perplexity.ai 的資料
- 存取您在 perplexity.ai 的資料
- 存取您在 grok.com 的資料
- 存取您在 copilot.microsoft.com 的資料
- 存取您在 www.google.com 的資料
選用的權限:
- 存取您在 chatgpt.com 的資料
- 存取您在 chat.openai.com 的資料
- 存取您在 claude.ai 的資料
- 存取您在 gemini.google.com 的資料
- 存取您在 www.perplexity.ai 的資料
- 存取您在 perplexity.ai 的資料
- 存取您在 grok.com 的資料
- 存取您在 copilot.microsoft.com 的資料
- 存取您在 www.google.com 的資料
- 存取您在 www.googleapis.com 的資料
- 存取您在 api.llmemory.xyz 的資料
開發者聲稱必須收集下列資料:
- 驗證資訊
- 網站內容
更多資訊