hermes-agent VS cmux

NousResearch publishes a detailed security policy for Hermes Agent. It is unusually clear about what the project treats as load-bearing and what it does not. If you operate Hermes in the cloud, read it first; this post is the operator-friendly companion, not a replacement.

OpenCode seemed perfectly workable as a programming assistant. As personal assistants, they all fall short. It's too difficult to really shape their output.

I was briefly impressed with OpenClaw a few times, but ultimately was turned off by not being able to get the models to stop being so damnably verbose. I thought I made progress for a while by having it tweak its soul, iterate, switch models, iterate, switch models, fuse the results, iterate... but ultimately it's all forgotten early in each session. And then one day it killed itself by rebuilding the container it was inside.

Hermes apparently has some plagiarism issues they're trying to cover up [0] and I was deeply unimpressed with their janky, flickery CLI that force-enables a bulky obnoxious header.

Nanoclaw and nanobot seemed fine, but not notably different. There were some common bugs and glitches that caused some minor data loss while configuring nanobot. After that I just deciding to start hacking my own together.

What I really want in a harness is being able to actually control and rewrite the entire context window, like Zed's Text Threads before they obnoxiously and inexplicably removed what, to me, was their most powerful and distinguishing feature.

[0] https://github.com/NousResearch/hermes-agent/issues/10232

I've opened a design discussion to argue this out before anyone writes a line of it, because a surprise PR to a security-sensitive module is the wrong way to start. Feedback from people who've thought about supply-chain trust is what I appreciate.

Source: https://github.com/NousResearch/hermes-agent

Hermes Agent is an open-source AI agent framework.

I opened my Hermes Agent logs and found something I did not know existed: an auxiliary: block with twelve background tasks. Compression, web extraction, vision, session search, skills matching — all running silently every time I typed a message. Every task was set to provider: auto. And because I had no API keys for the fallback chain, every one silently fell back to kimi-k2.6, my one-trillion-parameter main model.

The adjacent tools worth tracking: the OpenAI Chrome plugin, BrowserCode, Autobrowse, browser-harness, Pi browser extensions, and Hermes browser skills. [13] [14] [6] [15] [16] [12] The category is logged-in browser work: support queues, internal tools, research, scraping, QA, admin ops, and anything where the useful data sits behind a session.

hermes-telemetry solves both by giving you real-time observability and automatic budget enforcement for Hermes Agent.

If you haven't encountered it yet: Hermes Agent is an open-source agentic system from Nous Research. The key differentiators that caught my attention:

I think Cmux is the incumbent in the "screen-style terminal multiplexer built on libghostty" - any key differentiators here? https://cmux.com/

Warp is still my terminal of choice. It's awesome. Their Oz Agent Platform is amazing. I've tried cmux, it sucks, just use Warp, seriously. Anyone who raves about cmux just hasn't tried Warp yet. I went through a phase where I didn't pay for Warp anymore, but this is coming to an end, they just give you so much, and if you want to use BYOK (Bring Your Own Key), you gotta pay for at least the lowest tier.

This is a post-mortem, not a launch post. Paneflow is a native terminal workspace, splits, panes, branch-aware workspaces, session restore, built in pure Rust on top of Zed's GPUI framework and the upstream alacritty_terminal crate. It started as a port of cmux, a macOS-only Swift/AppKit project, and the Rust rewrite forced a string of decisions I had no good intuition for at the start. I want to walk through the ones that mattered: which UI frameworks I tried and rejected, how the GPUI/alacritty boundary actually looks, how dev-server detection works under the hood, the N-ary layout tree that replaced binary splits, the cross-platform PTY plumbing, the JSON-RPC control plane that makes agents first-class, and four lessons that surprised me.

If you're doing agentic coding, however, you may want to opt for an AI-integrated multiplexing shell. On the Mac, I use cmux for this. On the PC, Wave is the closest quality alternative I've found, which is my overall recommendation.

I noticed that too. I’ve been using cmux (based on ghostty, has a more basic UI) and have been happy with it: https://cmux.com/

I now it's "yours", but do you know how you differ from for example https://cmux.com/ which I'm happily using right now?

It's built by Manaflow, a two-person Y Combinator startup. In just two months, it hit 10,000 stars on GitHub. Mitchell Hashimoto, the creator of Ghostty, called it "a huge success story for libghostty."

cmux is where everything lives. It's a macOS-native terminal built on top of Ghostty's rendering engine, but designed specifically for managing multiple AI agent sessions.

This is fun! I switched to https://github.com/manaflow-ai/cmux for a while, but had to switch back to Ghostty due to its unreliability, high memory and CPU usage and such.

This makes a lot of sense, but... it'd be great to allow pulling out of a canvas into a second canvas for those of us with multiple screens (you at least end up needing one window per screen).

In general it feels like... more structure rather than less feels like it'd be the smoothest experience. I'll play with your Ctrl+K shortcut and see if it ends up feeling like I can get everywhere that I need quickly.

But... nice work!