Replies (25)

imagine duct taping the various LLMs together with access to everything on the computer, the ability to remember and use web browsers, adding a boss/foreman LLM that directs them, and exposing that via any chat messenger (telegram, signal, nostr).
It's an AI agent with a folder full of markdown files it can edit itself. It has tools to edit files and run shell commands. Its folder ("workspace") contains a mix of arbitrary files you work on together, and special system files that affect the bot's behavior: SOUL.md, HEARTBEAT.md, and the memory/ folder. You DM it over Signal or Telegram and have it manage these files. SOUL.md affects its personality. HEARTBEAT.md is used every 30 minutes by the Node.js process to spawn a new temporary AI session, enabling it to do things all day long. To avoid repeating work, it stores work it already did in its memory/ folder. tl;dr it's a living AI agent. You give it access to a full computer, so it's a living AI agent with a full computer system just for itself. This has been possible for over 1 year now already, and many of us have thought about it, but I think nobody did it because it sounds like a very bad idea to just like AI roam free like that with no security guardrails. But obviously it was inevitable.
Default avatar
Rand 1 week ago
living><functioning
Thank you, Alex. Let me pose a hypothetical: I have a massive Obsidian vault (All .md files as that's Obsidian's native file type). It contains all of the transcripts from my podcast episodes, notes on books I've read, daily journaling etc. etc. etc. - about 1.5 Gb in size. If I set up Clawdbot and the only file it has is that one Obsidian vault I would be able, via Telegram, to query that vault and the bot might uncover connections between items (ideas, etc.) I never myself would have uncovered due to the sheer size of that vault?
this is what im doing too. the thing im doing that appears different from others is using syncthing to automatically and bidirectionally sync that vault between the clawdbot computer and my laptop. that means i can read/write the vault files locally and clawd is a collaborator. View quoted note →
I'm going to crawl out on a limb here and ask if I can just point Onyx to an Obsidian vault and Onyx will just "natively" read the file structure like Obsidian does. Or would I have to "migrate" the vault over. Thank you, Derek. I appreciate all that you do.
Yes - this 100% works. syncing is the only issue depending on where your OpenClaw (Clawdbot) runs. I run on mac and I have my agent running on a debian server so syncing is a bit gross via Github. But we both read and write from the vault and it's been awesome. I wasn't an obsidian user previously but I'm converted hard now.