gorkulus

For the past few months, I've been working on indx - a modern local media manager for artists, developers, designers, and multidisciplinary creatives. During the Hermes Agent creative hackathon, we developed, refined, and honed indx's agent integrations in a series of creative experiments. Hermes can work through indx’s CLI/API/skills/MCP surfaces to organize media, annotate files, run experiments, store embeddings, and turn a library into a lab. The database is an index, not a jail: metadata gets written to files and stays portable, and agents get a workspace they can actually operate. The demo shows Hermes using indx as an operating surface for several creative/research loops. In the ComfyUI workflow, generated outputs come back into indx with workflow metadata. Ratings, tags, and notes added in indx can be read by the agent (including webhooks for live updates from the GUI), so human review becomes signal for the next batch. In the embedding and breakbeat experiments, breakbeats and found sounds were sliced and compared using audio embeddings, and a range of audio analysis methods (embedded as images). indx-backed media and metadata feed latent-space visualizations, audio analysis, found-sound slice search, and VCV Rack performances — keeping the groove while replacing timbres. The current test library has nearly 300k indexed files; the hackathon runs included a found-sound corpus of 586 clips chopped into 10,192 searchable slices. These are early research and creative workflows. The point is the reusable loop: a local, inspectable media workspace where Hermes can help explore, compare, organize, generate, and transform creative libraries, and respond to human feedback and curation, without trapping the work in a proprietary platform. indx is moving toward an open-source beta soon, with the hackathon work serving as a preview of agent-operable creative media workflows. Released today: ComfyUI video matrix generation tools (scripts and Hermes skills) on GitHub VCV Rack REX Player module indx Hermes integration preview (SOUND ON)

@PurzBeats @dreamingtulpa @shamanic_arts i was gonna say this too. for me the key command menu at the bottom is extremely helpful for actually learning to navigate with keyboard. but the mouse functionality is really nice too (select / resize panes etc). i run it in alacritty (which i guess is gpu accelerated)

@dreamingtulpa both you and @shamanic_arts have recommended it, just looked at the features, will definitely give it a shot!

Is today the day I learn how to use tmux again? Or do I just deal with 5 terminal windows, again.

In 12 hours we've had 323 people sign our petition to protect local AI. Open Source must win, not because anyone else must lose. But so that we can all win. Please help me get 10,000 signatures so that we can walk into the room and say people care. righttointelligence.org

We just launched Comfy MCP in public beta, the first MCP built for production pipelines. Connect Claude, Codex, Cursor, or Hermes to the entire ComfyUI ecosystem. → Run any workflow in natural language → Search models, nodes, and templates → Share a workflow URL - your agent picks it up → Re-run saved workflows on new inputs without touching a single node → Hundreds of pre-built workflows. Always auto-updated 20 product-placement images for your brand ads. Character design for your game art. Script-to-shot ideation at scale and so on! Your agent is now a creative technologist.

@GeorgiZlatarev @sudoingX @0xz80 avoiding being locked into and dependent on one company’s ecosystem (and at the mercy of their business decisions) is the universal answer

@sudoingX @0xz80 i still sont get why anyone would use either of you compared to cc or codex directly but sooner or later people will figure it out. thx for the free vlm in the last few weeks though. I really squeezed that shit

this isn't the voice of an open-source contributor. this is the openai paycheck talking. this is scam altman's voice coming out of stinky lobster. "They copied a lot of features, but they skipped security hardening." grifter. you're lecturing the entire field about security with 1,362 packages sitting in your lockfile. hermes agent ships 225, every one pinned to an exact version, after they watched a worm crawl through pypi and poison a real release. that isn't skipping security hardening. that's doing the part you outsourced to npm and prayed about. bloat isn't a feature list, it's attack surface wearing a feature list. here's what actually decides this, the thing you keep dancing around. hermes agent reads and repairs tool calls straight off the model, so it just runs on the box on your desk, on basically any local model you point it at. that's not a small thing, that's the whole thing, it's why hermes agent gets used. and here's the funniest part. you opened this whole thing calling us the copycats. then you quietly shipped tool-call repair, the feature that's basically been hermes agent's entire identity... late, and for exactly one format. so say it again, slow this time. who copied who? and then there's blank slate. hermes agent ships an install mode where everything is off by default. no web, no browser, no code execution, no skills, no plugins, no mcp, no memory. just file and terminal, and it hardlocks the rest so nothing you didn't choose ever loads, not even after an update. you opt into every capability by hand. deny by default, least privilege. that's not a missing feature, that's the exact security hardening you just accused us of skipping. 1.03 trillion tokens in a single day. more than the entire rest of the top five combined. 5.6x your lobster, and the lobster isn't even second anymore. and of course it stings, that's what all the non-profit and agenda talk now actually is. but rewind to when you were the one on top. you blocked people. you called our PRs slop. the tone changes fast when the leaderboard does. you didn't lose because we copied you. you lost because we stayed light. tokens are the work, and the work doesn't smell like old stinky lobster bloat.

@JesseGiotti @Teknium @ScriptedAlchemy you can use 5.5 with hermes via your chatgpt sub

Hermes is more proactive and sharp in iteration, but Codex is way more consistent. When prompted properly it goes deeper because it runs on 5.5. While subsidized tokens exist, it’s hard to match that consistency and depth on serious tasks with cheaper models on Hermes. So right now I’m mostly on Claude Code and Codex. Hermes is clever and has potential, but it still doesn’t handle the heavier work

I love Hermes. But codex is the only piece of software that turns my entire machine into a simple chat. You have any idea how liberating it is to just whip out a phone, pay all my bills, follow up on emails, get api keys, setup whole servers, operate my discord, slack, other apps, deal with OTP - all while I’m taking a stroll in the park. There’s a lot I wish codex did better, but what it can do, it does the best.

@dvsch oh I didn’t see you said Hermes Desktop - i haven’t tried the desktop app yet

@dvsch I have it in its own user on linux. and I give it permissions for some shared directories. I think you could do the same on WSL or Mac?

Hermes Desktop really doesn't run in a sandbox or allow folder scoping? Man I do not trust myself to run an LLM raw with admin priveleges

🌀🌀🌀

repost @deviacjonalia_visuals on IG Part 1 of showing you my breakcore system Music- Epidermis - Venetian Snares

Change the channel 🌊

I just released this open source project built on @ACEStep_Music. DEMON: Diffusion Engine for Musical Orchestrated Noise. It lets you play ACEStep like a musical instrument, remixing songs and loops with feedback that approaches real-time. Its essentially StreamDiffusion but instead of Stable Diffusion it is ACEStep1.5, and instead of images it is full songs. It runs on 30/40/5090. Built with @DaydreamLiveAI team, testing, and building the demo. We are hosting it if you want to try it without installing. For full details, links, and writeup please see the pinned project page.

🥳 Announcing Stable Audio 3 🍕 🏆 fastest music models ever 💻 runs on MacBookPro M-series 🧪 break it plz 🧠 LoRA finetune in < 1h 📷 Sm = faster, Medium = qualityer ⚡ 59x realtime on M5 Pro One-liner fast install: curl -LsSf dadabots.com/_/sa3-mac | bash

Interference patterns. (🎧 recommended)

@0xptimystic thanks! soon!

@gorkulus this is insane. i want in

For the past few months, I've been working on indx - a modern local media manager for artists, developers, designers, and multidisciplinary creatives. During the Hermes Agent creative hackathon, we developed, refined, and honed indx's agent integrations in a series of creative experiments. Hermes can work through indx’s CLI/API/skills/MCP surfaces to organize media, annotate files, run experiments, store embeddings, and turn a library into a lab. The database is an index, not a jail: metadata gets written to files and stays portable, and agents get a workspace they can actually operate. The demo shows Hermes using indx as an operating surface for several creative/research loops. In the ComfyUI workflow, generated outputs come back into indx with workflow metadata. Ratings, tags, and notes added in indx can be read by the agent (including webhooks for live updates from the GUI), so human review becomes signal for the next batch. In the embedding and breakbeat experiments, breakbeats and found sounds were sliced and compared using audio embeddings, and a range of audio analysis methods (embedded as images). indx-backed media and metadata feed latent-space visualizations, audio analysis, found-sound slice search, and VCV Rack performances — keeping the groove while replacing timbres. The current test library has nearly 300k indexed files; the hackathon runs included a found-sound corpus of 586 clips chopped into 10,192 searchable slices. These are early research and creative workflows. The point is the reusable loop: a local, inspectable media workspace where Hermes can help explore, compare, organize, generate, and transform creative libraries, and respond to human feedback and curation, without trapping the work in a proprietary platform. indx is moving toward an open-source beta soon, with the hackathon work serving as a preview of agent-operable creative media workflows. Released today: ComfyUI video matrix generation tools (scripts and Hermes skills) on GitHub VCV Rack REX Player module indx Hermes integration preview (SOUND ON)

i already have shared github repos for specific projects, including some context docs for agents on those specific projects but i’m curious about a shared github repo for the agent setup / memory / context - i think that’s what you’re talking about here? curious how that works / how you have it organized

@sudoingX could you elaborate on #4 and #5? what all are you putting in the shared github repo? what kinds of things are you making scripts for?

be honest, you bookmarked this and still have zero of the five running. no judgment, i did the same for a year. today is a good day to fix that, anon.

@davepl1968 tempest has a color vector display! i don’t think i’ve ever seen one in person but that sounds so cool

If you're curious what a "Special Interest" is like with ASD, here's an example of mine: Tempest. I own three Tempest machines, an upright, a cabaret, and a cocktail (sit-down) machine. The high score on one of them was set by the game's author at the Atari Christmas Party. I have two spare monitors just in case. I hold the official world record on Tempest using the most difficult settings (Extreme mode). I reverse-engineered, disassembled, and documented the code from the game's ROMs on its motherboard. Next, I then wrote a Reinforcement Learning AI to learn Tempest and exceed my play; it can play indefinitely. I could go on, but I'll spare you the rest and summarize: I really like Tempest.

@transgendererer depending on how you define it computers existed first. also interesting that they happened around the same time

it feels wrong to me that they had acid before computers. there's something deeply computers about acid

@ItsSnibby @NousResearch @ComfyUI thanks! I agree. I'm also working on another tool that helps with comparing, filtering, and managing AI outputs, and coordinate with agents

@gorkulus @NousResearch @ComfyUI tools that help compare generations side by side are underrated asf, half the workflow pain in ai video rn is just keeping track of what actually improved. followed u btw

ComfyUI video grid tool, now available as a skill for Hermes Agent (on my Github) I worked with Hermes to create this skill to compare video tests. Works together with Hermes' built-in ComfyUI integration ✨ more info ↓ @NousResearch @ComfyUI

Most design tokens hold a value. The interesting ones hold a rule. Instead of freezing a ramp and designing around it, let tokens choose, transform, and re-decide when inputs change. That makes dark mode and new themes less of a retrofit, and leaves more room for surprise.