正在加载视频...
视频加载失败
Today's Hermes Agent Masterclass is the finale! Module 10 covers security, an important topic for running agents to ensure each agent can perform its tasks while minimizing exposure. Hermes has a ton of built-in security features. In this clip, I show how you can set approvals for each profile... show more
21 条评论

Full YouTube link: Hermes Agent Masterclass: 10. Security

I’ve been sharing your series with several people. Thanks for putting this all together

Thank you! I hope it helps!

You my friend, deserve an award from @NousResearch (or at least a kiss from @Teknium 😂)

Hermes also thrives with ContextLattice

Are there transcripts available for these videos that I can feed into Hermes?

Tonbi the goat with another banger 🔥 Well done, this whole series is top top top notch

how do you usually decide between smart approvals vs YOLO mode when running longer agent sessions?

All ten of these modules are gold! Highly recommended if you are a Hermes Agent user.

watched approvals pass while agents still hid bad intent in tool args

Do you let Hermes loose on your email?

Thank you boss✨✨

does the approval flow add much latency in practice?

Can i be the first 2 session vive coding client ?

Hermes agent has great list of identifying and blocking those severe commands but still terrifying to run it for root folder.

But i can always ask to it tell wipe my file system just like grandma used to do it. B

This is cool I'll add into my Hermes

Crack, gracias

Something I've had Hermes do before, is scold me about putting any passwords into Telegram chat to turn around and read one from the env file and post it as a Terminal dialogue in telegram chat. @Teknium . I didn't notice it at first but later did.

Great stuff. Just stumbled across it via @Teknium and loving it. Great voice! Is that your real one or AI?

How was your experience using Grok 4.5 with Hermes compared to GPT 5.5?
