Hars
157 posts
@eliana_jordan not too late at all. the real fun starts after install when you figure out what should be agentic vs what should stay boring and deterministic.
English
you know what
all of these "which is better" polls are silly
use codex or claude code, whatever works best for you
i am grateful we live in a time with such amazing tools, and grateful there is a choice
English
@NousResearch @Shopify this is the interesting part imo. not agents as cute demos, agents plugged into real ops like inventory, orders, fulfillment. commerce is such an obvious wedge for this.
English
Shopify is the all-in-one commerce platform powering millions of businesses worldwide
Thank you to the @Shopify team for building their own official Hermes Agent skill enabling your agent to manage products, orders, inventory, and fulfillments from any channel.
English
@outsource_ This has always eluded me. How does one create and breakdown a project in a way that it can be tackled by a swarm of agents?
English
HermesAgent SWARM is HERE Hermes-Workspace.com
🚨 NEW RELEASE v2.1
> Unlimited Hermes Agents
> 1 Orchestrator
> 0 Humans
💻 Orchestrator Chat
🤖 Multi-Agent Control Plane
🏆 Kanban TaskBoard
📄 Reports + Inbox
⌨️ TUI View, built in
English
@mattyp @NousResearch Made 2 PRs that were just merged:
PR 1, LLMs.txt: github.com/NousResearch/h…
PR 2, better navigation:
github.com/NousResearch/h…
English
. @NousResearch & @Teknium a very simple way to get more people using Hermes is to spend some time on your docs:
- Agent unfriendly (No llms.txt / llms-full.txt)
- Unclear product story
- No deployment path
- Overwhelming navigation structure
Could be 1000x easier to get started
English
Hars retweetledi
Your team is not slow.
Your business info is scattered.
Product details in folders.
Pricing rules in someone’s head.
Customer history in email.
Stock updates in WhatsApp.
Specs in old PDFs.
That is how simple work becomes slow work. #buildingmaterials #automation
#b2b
English
@robinebers Just press escape
Toggle plan mode off
Switch model or reasoning effort
Then say implement plan
Hit enter
Voila
I'm not missing smth right
English
In Codex, I wish there was a model selector before you accept a plan.
feels like planning with High/Extra High and implementing with Medium is pretty common
English
English
I have to go out of town for a funeral thru the weekend but I am leaving everyone with one new cool feature inspired by ralph loops and Codex's upcoming /goal feature.
If you use /goal , it will start a loop with a supervisor model determining whether the task completed at the end of an agent loop - if it hasn't it will force it to keep going until it's done!
Enjoy and have a great weekend.
PR: github.com/NousResearch/h…
English
Hars retweetledi
Hars retweetledi
Most people I’ve met doing 7–10 figures were not doing anything genius.
They picked an industry, stayed in it for years and became known.
That created steady bread-and-butter income.
Then their reputation and connections gave them access to investments most people never see.
English
This is what I'm working on with @Dan1sh_Kh
AI audits for the industrial/material supplier industry, starting off timber focused
Trying to integrate AI into this space among other things
scaleOS@scaleosuk
A sent quote is not a managed opportunity. No owner. No follow-up date. No buyer status. No visibility. That quote is already at risk. 2 missed orders × $3k × 18% margin = $1,080/month in potential margin leakage. Before chasing more leads, fix how existing demand is handled.
English
It's only been two months, can't believe the growth and community we've built, together, around this!
Thanks again everyone.
Nous Research@NousResearch
Meet Hermes Agent, the open source agent that grows with you. Hermes Agent remembers what it learns and gets more capable over time, with a multi-level memory system and persistent dedicated machine access.
English
cuz everyone likes not using their brain and wanna be spoonfed😂
Danish@Dan1sh_Kh
Why don’t people who make AI agent skills also evaluate them against trigger boundaries, response contracts, mode discipline, failure recovery, scope control, reference quality, and output consistency? And when it makes sense, include a runnable eval harness too?
English