DataMoat
Language: English | Português (Brasil) | 简体中文 | 繁體中文 | 日本語 | 한국어 | Türkçe | Русский | Tiếng Việt | ไทย | Deutsch
Official website: https://datamoat.org
GitHub repo: https://github.com/max-ng/datamoat
Protect, export, back up, analyze, search, and reuse everything you build with ChatGPT, Claude, Codex, Cursor, DeepSeek, Qwen, and OpenClaw
Local encrypted backup archive for sessions, images, files/PDFs, and SKILL.md folders.
DataMoat keeps your AI work history local and encrypted, preserving raw source records intact and building a normalized layer for analysis, search, export, reuse, handoff, and private AI memory.
Your most valuable future AI data is already disappearing.
Download DataMoat now to see how much ChatGPT, Claude, Codex, Cursor, DeepSeek, Qwen, and OpenClaw work history you can still capture.
Core backup scope: DataMoat backs up supported skills + sessions + attachments into the same encrypted local memory archive. Skills are saved as full folder snapshots, not just names.
The people and companies that own their AI data will win the future.
DataMoat is an AI work history memory archive for people and teams working across ChatGPT exports, Claude CLI, Claude Desktop, Codex CLI, Codex app, Cursor, DeepSeek and Qwen through Claude Code GUI workflows, OpenClaw, and other AI tools. It preserves the full working record: sessions, locally stored thinking tokens and reasoning blocks when present, prompts, responses, tool output, files, attachments, metadata, skills folder contents, and original source records on the same machine, so your work stays reviewable, protected, reusable, and easier to hand off later.
How DataMoat Stores Your Work
DataMoat keeps two layers:
- Raw archive: original session JSONL, SQLite records, logs, attachments, metadata, skills folder snapshots, and any locally stored thinking tokens or reasoning blocks are preserved as close to the source format as possible.
- Normalized index: records from different tools are converted into a common schema so you can analyze, search, review, export, reuse, and hand off work across tools.
Supported sources today: ChatGPT export ZIP/folder imports, Claude CLI, Codex CLI, Codex app local sessions, Claude Desktop local-agent sessions on macOS, supported local Cursor agent transcripts, DeepSeek and Qwen sessions when written locally by Claude Code GUI workflows, and supported local OpenClaw session records.
More data sources and platform releases are on the roadmap: star and watch this repository so you can follow new capture integrations and platform updates as they ship.
New In 2.0.7: ChatGPT Export Memory Import
DataMoat now imports supported ChatGPT export ZIP files or extracted export folders into the same encrypted local memory archive used for Claude, Codex, Cursor, DeepSeek, Qwen, OpenClaw, skills, and attachments.
