ทวีตที่ปักหมุด
This is exactly what it feels like wielding a powerful open-source agent harness… pure magic.
English
Agent Empire
741 posts
Agent Empire
@ForgeAgents
#OpenSource #HermesAgent #OpenClaw, CLI, AI. Forging Agents All Day, EveryDay! 🚀
Ontario, Canada เข้าร่วม Şubat 2026
60 กำลังติดตาม200 ผู้ติดตาม
@AlexFinn The entire history of humanity there has been opportunities!
Life is about choices.
x.com/ForgeAgents/st…
Agent Empire@ForgeAgents
Marc Andreessen is always a great listen. There will be more jobs/tasks from ai than any other time in history.
English
Universal Basic Income would be a massive mistake
It would tear apart the fabric of society. Humans need purpose. They need agency
If you hand them all free cash they'll be happy when they receive the check, but will quickly become depressed
We need Universal Basic Opportunity
EVERYONE needs the opportunity to build anything they want. Any business they want. Any product they want
If you have the opportunity to build anything you want and still squander it, that's on YOU. YOU face the consequences of your inaction
The good news is AI creates UBO. With AI you can create quite literally anything you want.
I did. I started one successful business and I'm about to launch another. I've never been happier or more satisfied in my life. Could never do it without AI
A free pro tier version of ChatGPT for every citizen would be WAY more productive and fulfilling for a majority of society than a free check.
Deep down people don't want free crap. They want equal opportunity.
English
whats your favorite model to use with Hermes agent?
English
Some new improvements to performance just went in.
Python gets a bad wrap for performance but we aint looking to shabby against a trillion dollar co's rust codebase, beating codex at most multi-turn tasks we benchmarked (mt stands for multiturn)
PR: github.com/NousResearch/h…
English
@TansuYegen Life is about choices. 😎
Choose wisely.
x.com/ForgeAgents/st…
Agent Empire@ForgeAgents
This is exactly what it feels like wielding a powerful open-source agent harness… pure magic.
English
Of course. My point is these problems will be mitigated and are being mitigated as we speak. Everyday, especially with Hermes things are getting tighter and more roubust.
Our job is to hammer these technologies like our life depends on it. Stretch and test it vigorously.
The first 2 months with openclaw I babied it, worried about prompt injections, setting up perfect config files etc.. etc..
and
got very little done.
When i released it, whole different world.
Now with hermes, WHOLE DIFFERENT UNIVERSE!
😎
English
@ForgeAgents @AlexFinn I know Hermes. But this is another context, and Hermes is also susceptible to attacks like any other agent.
English
You should NEVER be downloading AI agent skills from the internet
It's the biggest attack vector for security breaches right now
Tons of skills in the public skill sites you see are compromised with prompts that open your computer up
Do this instead:
If you see any skill you want your agent to have, give the link to your agent
Say "look at what this skill does. I want you to do something similar. Think about how this fits into our workflow and how you'd use it. Then build your own version of it, custom for what we do"
Your agent then will build its own, secure version that can't have dirty prompts from random people on the internet
This is the biggest security practice you can implement
English
I have been running bare metal, ZERO isolation for months now. It is magical. 😎
The future of opensouce agents/harness is the wild west and should stay that way.
Use Hermes! It has Built-in Protections (They’re Quite Good)
Hermes already has several layers against prompt injection:
Context file scanning
Dangerous command approval
Secret redaction and credential filtering.
Memory & skill scanning.
Plus you can prevent injections with guardrails.
x.com/ForgeAgents/st…
English
This is a huge mistake. If there's a prompt injection hidden in the files, the LLM will read the file, apply the injection to your system, and you will still think you did a 'good job,' which is not the case.
A proper way to follow your approach would be to manually inspect the skill, understand its capabilities, and then manually describe it to your AI model.
English
Agent Empire รีทวีตแล้ว
Bare metal, zero sandboxing/docker/safeguards. It has to be pushed like pioneers went bare metal in the wild west.
When I stopped trying to box it in using huge endless .md files and rules. Its been performing much better. Overwriting its own config with errors alot of the time due to me overwriting its soul and agent md files.
In this awesome interview Marc Andreessen (netscape) talks about how he has friends that run openclaw like it was the wild west. Getting their claws to control everything (full root access, giving them bank accounts, their smart home stuff etc..) with no safety.
He says it the only way it will progress, and people will be praised for doing so in these early days.
All Marcs writings and interviews are worth the time.
x.com/Claw_empire/st…
English
@AlexFinn A great LLM and hermes can build anything.
It's over, just has to sink in.
RIP apps and most software.
English
@mitchellh That is right. Opensource "FREE" is freedom!
There is forever ongoing income in making things bulletproof and usable.
x.com/ForgeAgents/st…
Agent Empire@ForgeAgents
When I first started working as a dev in my 20's at Bombardier Aerospace in late 90's. My boss and many others said the internet was never going to be much bigger than it was. In TLUG (Toronto Linux User Group) meets during that time, there would be 20 -30 people max at the meetings. People laughed their heads off at Linux in those days. x.com/Claw_empire/st…
English
This is why RedHat makes like over $6 billion a year by the way. For anyone wondering what the fuck RedHat does, its basically: take the free chips on the side of the road, validate them, and sell the supply chain contract.
English
Supply chain attacks and OSS sustainability go hand in hand. I've semi-seriously joked for years that OSS upstreams should periodically purposely inject full vulns into their code and let downstreams fuck around and find out. Downstreams can pay to get the non-FAFO version.
The not joke part is simply that OSS maintainers aren't a supply chain. OSS maintainers are not responsible for monitoring CVEs (because, they are not a supply chain). OSS maintainers are not at fault when bad shit happens to downstreams, because basically every OSS license (MIT, Apache, GPL, etc.) literally says: the software is provided "as-is, without warranty." You get what you pay for (that is to say: absolutely nothing!)
Now, the joke part is that I do believe there is an ethical obligation to try to prevent harm downstream. But "try" is the key word. So, this isn't a serious proposal.
But, if you're using OSS code and you're not paying for a license with a contract that promises some kind of warranty, you have no supply chain. You (the downstream user of an OSS lib) ARE the supply chain.
To use a metaphor: physical goods have a real supply chain. Car manufacturers, chips, clothes, toys, etc. You have a signed commercial agreement with all your suppliers that promises quantity AND quality and blowback if either are missed. Thats a supply chain.
If someone puts some chips on the side of the road with a "FREE" sign, then you integrate those into a product, then find out those chips are hacking customers, its your fault, not the person who dropped them on the side of the road.
English
@Teknium Bro, you guys shipped the most engaging product on Earth!
I just use Hermes and don't get 8 hrs of sleep.
English
finally made the switch to hermes.
i literally just said to codex "set up hermes. completely switch everything over from openclaw, keeping everything in tact. bind to the same discord server."
and it ran for like 5 min and just worked. no debugging.
i'm going to get ai psychosis from just being in constant awe at this stuff every single day. nothing feels real anymore.
English
