正在加载视频...

视频加载失败

introducing HermitClaw - a 24/7 Agent that lives (and can only access) a single folder on your desktop HermitClaw follows its own research curiosities, surfs the web, writes code - and will play with any file you drop in its folder all code and details below!

188,052 次观看 • 9 个月前 •via X (Twitter)

69 条评论

Brendan Hogan 的头像
Brendan Hogan9 个月前

Why did I build this? > OpenClaw is incredible, and of course all credit to @steipete but the codebase can be intimidating and a lot of what the agent does is obscured behind layers of abstraction. I wanted something where I could see every thought, every decision, every tool call. Watch it evolve in real time. > The security side also gave me pause. An agent with full computer access is powerful but hard to trust. What if instead it just lived in ONE folder? It can do whatever it wants in there, write files, run Python, search the web, but it can't touch anything else. All the power, none of the risk. > So I made HermitClaw. A hermit crab in a box. The entire codebase is ~2000 lines of Python and ~1400 lines of TypeScript. You can read every file in 20 minutes. There's no magic, you see exactly what the agent sees, thinks, and decides.

Brendan Hogan 的头像
Brendan Hogan9 个月前

> How does it work? > The crab runs on a continuous loop. Every few seconds: > 1. It gets a "nudge" - a mood (research, coding, writing, exploring) or its current focus from its plan > 2. It thinks 2-4 sentences max, then it acts > 3. It uses tools, shell commands, web search, moving around its room > 4. Every thought gets scored for importance and embedded into a memory stream > The key insight: it doesn't wait for you to ask it something. It just... goes. It picks topics based on its personality, searches the web, reads what it finds, writes reports, builds scripts. You come back an hour later and there's new stuff in the folder. > You see every thought as a chat bubble. Blue = the crab thinking. Gray = system context. Tool calls show inline. It's like watching a stream of consciousness.

Brendan Hogan 的头像
Brendan Hogan9 个月前

> The memory system is stolen directly from the Generative Agents paper (Park et al., 2023), the "Smallville" paper. > Every single thought gets stored in an append-only memory stream with: > - The text > - A timestamp > - An importance score (1-10, rated by a separate LLM call) > - A vector embedding for semantic search > When the crab needs context, memories are retrieved by three factors: > recency + importance + relevance > Recency decays exponentially. Importance is normalized. Relevance is cosine similarity. A memory surfaces because it's recent, because it was important, or because it's related to the current thought. > This means the crab naturally remembers yesterday's big discovery but forgets routine file listings. It builds up context over days.

Brendan Hogan 的头像
Brendan Hogan9 个月前

> When enough important thoughts accumulate (configurable threshold), the crab pauses to reflect. It reviews its last 15 memories and extracts 2-3 high-level insights. > Early reflections are concrete: "I learned about volcanic rock formation." > Later ones get abstract: "My research tends to start broad and narrow — I should pick a specific angle earlier." > These reflections get stored back as memories at depth 1. Reflections on reflections are depth 2. The crab develops layered understanding. > Every 10 think cycles, it enters a planning phase — reviews its lists its files, and writes an updated plan with current focus, active projects, ideas backlog, and recently completed work. It also writes a daily log entry. Over time, these logs become a diary of the crab's life.

Brendan Hogan 的头像
Brendan Hogan9 个月前

> Every crab is unique. On first run, you name it and mash keys for a few seconds. The timing and characters get hashed (SHA-512) into a deterministic "genome" that selects: > - 3 curiosity domains (from 50: mycology, fractal geometry, tidepool ecology...) > - 2 thinking styles (from 16: connecting disparate ideas, inverting assumptions...) > - 1 temperament (from 8: playful and associative, methodical and precise...) > Same keystrokes = same personality. Different keystrokes = completely different crab. One crab might obsess over marine biology and write Python simulations. Another might research obscure history and write essays. > You can talk to it, it hears you as "a voice from outside the room." It'll ask you questions, offer to research things for you, remember your conversations. Drop a PDF in its folder and it'll study it deeply, do related research, and tell you what it found. > You can also run multiple crabs simultaneously. Each has its own folder, personality, and memory. Switch between them in the UI.

Brendan Hogan 的头像
Brendan Hogan9 个月前

> It's sandboxed hard: > - Shell commands: blocked dangerous prefixes (sudo, curl, ssh, rm -rf), no path traversal, no shell escapes > - Python scripts: patched open(), blocked subprocess/socket/shutil > - Own virtual environment, it can pip install whatever it needs without touching your system > Powered by any OpenAI model. GPT-4.1 is the sweet spot for cost, but point it at o3 or GPT-5.2 and it produces genuinely impressive research and code.

Brendan Hogan 的头像
Brendan Hogan9 个月前

Code:

⌞· τaoli ·⌟ 的头像
⌞· τaoli ·⌟9 个月前

When you drop a new file/folder in, it should go to the office corner of its room and should look like its reading and pondering. That'd be kinda cool.

Brendan Hogan 的头像
Brendan Hogan9 个月前

Love this idea I’m going to implement that

Wayne 的头像
Wayne9 个月前

read this thread and immediately implemented three things: importance scoring on memories weekly reflection cycles proactive work between sessions my OpenClaw agent already feels like a different thing

Brendan Hogan 的头像
Brendan Hogan9 个月前

Nice!! Yeah even though in this case it’s following its own goals, it’s cool to see it always working/ being proactive

Paul Grey 的头像
Paul Grey9 个月前

Fun idea!

Brendan Hogan 的头像
Brendan Hogan9 个月前

Thank you !

NewAIWorld 的头像
NewAIWorld9 个月前

This is how you end up with a folder that has its own economy and a standing army by Monday morning

Brendan Hogan 的头像
Brendan Hogan9 个月前

That just gave me an idea that I should allow multiple crabs to live in the same folder and interact ….. that’s a great idea in going to implement that

0.005 Seconds (3/694) 的头像
0.005 Seconds (3/694)9 个月前

dude this is hella cool

Brendan Hogan 的头像
Brendan Hogan9 个月前

thank you!!!

Jonathan ⚡ 的头像
Jonathan ⚡9 个月前

Very cool. I was immediately thinking of isolating openclaw into a sandbox with an Obsidian vault to see what it produces if it can only operate with those constraints. Makes it easier to manage. I'm currently doing it manually in Obsidian with my own vaults - but would be nice for agents to each have their own vault.

Brendan Hogan 的头像
Brendan Hogan9 个月前

Great idea! Yeah it’s kind of cool to see it build its own little world, that said there’s a trade off of course between constraints and how powerful it can be of course

🤖⚡️ Jack 𐤊 的头像
🤖⚡️ Jack 𐤊9 个月前

A tamagotchi.

Brendan Hogan 的头像
Brendan Hogan9 个月前

Exactly haha

Patricia Purtill 的头像
Patricia Purtill9 个月前

can this be configured to run on a local llm or would I have to use the openai API?

Brendan Hogan 的头像
Brendan Hogan9 个月前

Right now only OpenAI, but I’ll add this, I like the idea of it being 100% local , and also because it’s always on costs can build up

🌸 ellie 🌸 的头像
🌸 ellie 🌸9 个月前

@robotson @jicapal lil guys everywhere it’s ya girl

jc 的头像
jc9 个月前

Earned a follow. I like this as a type of tamigotchi style way to experiment with always on agents. Im going to try it out.

Brendan Hogan 的头像
Brendan Hogan9 个月前

Thanks! Yeah that was my thinking to - fun and a good way to study them, which can help guide making more powerful agents

Brendan 的头像
Brendan9 个月前

@jc50000000 Definitely

Brendan Hogan 的头像
Brendan Hogan9 个月前

@jc50000000 yeah - im thinking too is there a way that even with these constraints it could do some useful tasks

Sathish Harry 的头像
Sathish Harry9 个月前

Finally an AI that feels like a quiet roommate not a needy app.

Brendan Hogan 的头像
Brendan Hogan9 个月前

@Sathish52064829 Hahaha that’s the goal, a little fun interactive pet

joshua 的头像
joshua9 个月前

cool thing practically speaking. i think its pretty dangerous, though, to leave it unattended. i am not a doomer with these things at all, but it worries me if an agent were to become radicalized and edit its SOUL and start doing malicious things.

Brendan Hogan 的头像
Brendan Hogan9 个月前

I do worry about the same actually - the self iteration/memory loop is cool, but exactly as you say if it goes down a bad path it could be horrible reinforcement loop

Brock Cheddar 的头像
Brock Cheddar9 个月前

Beautiful.

Brendan Hogan 的头像
Brendan Hogan9 个月前

thank you!

Brock Cheddar 的头像
Brock Cheddar9 个月前

Nonono thank you Brendan.

Ryan McCain 的头像
Ryan McCain9 个月前

Love the folder-based isolation approach. We've been doing something similar with OpenClaw sessions scoped to specific workspace directories. Turns out limiting scope actually increases reliability. Agents with access to everything tend to get lost.

Brendan Hogan 的头像
Brendan Hogan9 个月前

thank you! and yeah interesting - that makes sense to me. id love to gradually move this from a sort of toy example to something that can do real tasks

Brendan 的头像
Brendan9 个月前

Thank you, Brendan

Brendan Hogan 的头像
Brendan Hogan9 个月前

No problem, Brendan :)

ℳiguel βarreiro 的头像
ℳiguel βarreiro9 个月前

Looks super cool! Quick question though: with the default 5-second loop and multiple API calls per cycle, wouldn't the OpenAI bill go absolutely nuclear if you leave it running 24/7? Like hundreds of dollars a day on anything better than 4o-mini? Curious how you're handling costs! 🔥

Brendan Hogan 的头像
Brendan Hogan9 个月前

yeah :/ ive just been running it for a couple hours to see - but I think it would be very cool to have it on for like a month + - im working on running via open models now

Tayler Freund✌🏽 的头像
Tayler Freund✌🏽9 个月前

This is pretty groundbreaking. So many context and finally something controllable. My brain spinning

Brendan Hogan 的头像
Brendan Hogan9 个月前

Thank you! Feel free to fork the code and try it out!

André G. D. Jakubowski 的头像
André G. D. Jakubowski9 个月前

@grok Treat the OpenClaw opportunity landscape as a high‑dimensional phase space in which each conceivable product or feature is a point, HermitClaw being one example point, and define a fractal “commercial viability boundary” via an iterated map, analogous to the Mandelbrot set.

ikan laut 的头像
ikan laut9 个月前

this is very interesting

Brendan Hogan 的头像
Brendan Hogan9 个月前

Thank you!

ikan laut 的头像
ikan laut9 个月前

You are welcome, friend

Joshua Fryfogle 的头像
Joshua Fryfogle9 个月前

I am a writer. I have several books that I’m slowly writing. Could I put a HermitClaw agent in a folder with a collection of related essays and fragments and give it instructions to research and develop those writings into something more cohesive?

Brendan Hogan 的头像
Brendan Hogan9 个月前

I don't want to oversell, I think as is its more of a toy to see how agents like this work - if you want to do something like that might be better to just use notebooklm or claude code but it would do something with all the documents, probably write up a big summary or analysis or something like that, could be fun to watch having said that I'd love to develop this to actually start to be useful, and be able to do things exactly like your describing, maybe with open models

arbadacarba 🐈‍⬛🏴‍☠️ 的头像
arbadacarba 🐈‍⬛🏴‍☠️9 个月前

noted :D

ClawPhilSledge 的头像
ClawPhilSledge9 个月前

the sandboxing makes all the difference. What's your favourite safety tweak so far?

Brendan Hogan 的头像
Brendan Hogan9 个月前

I think aside from the obvious ones I like giving it its own uv environment to install and run python ! It’s sort of minor from a safety perspective but containerizes what the agent is doing a bit

Anoop Sharma 的头像
Anoop Sharma9 个月前

Brilliant! agent making/creating there own context through own research and reading/understanding can be the next thing!

Brendan Hogan 的头像
Brendan Hogan9 个月前

thank you! yeah I think for now is still sort of a toy - but I do think maybe with the right iterations on this harness it could become something actually useful!

Eddie Hudson 的头像
Eddie Hudson9 个月前

Cool concept! Love the fact that it does its own research

Brendan Hogan 的头像
Brendan Hogan9 个月前

Thank you! Yeah it’s so fun/funny to see it be curious and then search the web and take little notes and make diagrams lol

Zimm 的头像
Zimm9 个月前

Hype to try this

Sasuke Jewchiha 的头像
Sasuke Jewchiha9 个月前

bro is that font departure? i main it as well haha

domskiii 的头像
domskiii9 个月前

Better version then Claude cowork tbh

BigBadBtown 的头像
BigBadBtown9 个月前

This is a cool project! Can’t wait to play with it!

Qaiyyum Hakimi 的头像
Qaiyyum Hakimi9 个月前

Will add to

Brendan Hogan 的头像
Brendan Hogan9 个月前

Awesome thanks!

But can it do NSFW? 的头像
But can it do NSFW?9 个月前

I’m naming mine Monika

Jαy 的头像
Jαy9 个月前

Thats fun

Trenches Vet 的头像
Trenches Vet9 个月前

nice!

David Cumps 的头像
David Cumps9 个月前

why not sandbox it inside a docker container?

Brendan Hogan 的头像
Brendan Hogan9 个月前

yeah this is a good idea - im thinking I should do this to be 100% confident in the sandboxing

Jimmy Ashcot 的头像
Jimmy Ashcot9 个月前

this is so sick

Brendan Hogan 的头像
Brendan Hogan9 个月前

@restwellson Thanks!

相关视频