Hacker Newsnew | past | comments | ask | show | jobs | submit | practicalsystem's commentslogin

Pretty happy with the experience of using it so far. Asked it to do something to blow me away and it made a podcast about the features it made while onboarding itself to my opensource project called DashClaw.

Then I had it make a marketing video as a test, pretty neat. Obviously all of this was doable but quite impressed with the model underneath since it's all free, for now.

https://x.com/ucsandman/status/2097498962714341661


tldr;

private data + untrusted content + outbound capability = hard stop

a markdown file is a wishlist, real security lies within hooks and plugins


it's wild how far behind society is to current LLM capabilities

cool going to check it out, I've got an opensource project that might compliment it that I'm excited to try

https://github.com/ucsandman/declick


harness engineering will be a job in the future imo

it is an interesting field to watch out. OpenAI has started this journey, lets see how this shape up. FDE roles were already getting some flame this week. May be everyone now will change the name or shift to Harness engineering :)

I wish they would just call us problem solvers and be done with it, that's all it's ever been and ever will be

that is pretty accurate. when FDE roles started popping up and i looked at the description, i was not sure how is it different that a typical implementation role. everyone needs some fancy words for showing future growth.

great idea, excited to test it out between my openclaw agents, claude, and codex

glad to hear! feel free to get in touch on our discord if you think of some areas for improvements or to share feedback. https://discord.gg/B3f8QjqeBa

is there a way to load my current agents memory because I don't want to start over fresh

Great idea, going to look into that soon.

excited to test it, great idea

I had a similar idea for an opensource video editor where you edit along side the LLM but I gave up on it haha


Glad you like the idea. Right now is only an editorial step for automatic video processing/editing workflows. But I will love to hear what ideas you have for: editing alongside LLMs.

https://github.com/ucsandman/Magnetic

honestly this was in June which feels like forever ago and I'm not entirely sure what my plan was haha. I knew I was going to be doing marketing for my projects and thought an editor with LLM capabilities would make it easier, but then I just made a repo and a skill where agents can make marketing material directly

https://github.com/ucsandman/marketing-studio


Thanks for sharing.

Looking at your projects brought a couple things to my mind. I use the agent to make the initial cuts removing silence, etc, from the transcript; so when the project is loaded it already is "half" baked. Then I just need to adjust the cut.

Do have a similar marketing workspace for my projects, one for each, with a abstract repo that have all skills and brand assets, tone of voice etc. By starting a new project the new workspace initialize by interviewing the user to gain all information of the project/product and then start defining assets, tone of voice, workflows, etc. Been using that for several month now, it gets better as the models get better.

Wondering what I need to adjust to open source it...


thanks! would love some feedback if you have any

going to have claude go through all these posts and update it's harness, hilarious idea for a site but sneaky useful I think

great timing with codex, claude, cursor, and grok being down

long live opensource


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: