Notes on Cross-Platform Computer Use Agents for macOS and Windows — Selection with Hermes Agent and cua-driver as of September 2026

Notes from comparing Computer Use agents that can operate native apps on both macOS and Windows as of September 2026. Among official apps, only the ChatGPT and Claude desktop apps clear the bar of two-OS support plus full-OS control. On the open-source side, Nous Research's Hermes Agent paired with trycua's cua-driver is the strongest starting point, with UI-TARS, Agent S, and Open Interpreter as adjacent candidates. This article focuses on what feature-matrix comparisons tend to hide: the actual scope of the 'no-foreground contract' for background operation, TCC and Accessibility on macOS, UIPI and Session 0 on Windows, multi-monitor limits, and Japanese IME behavior.

HTTP 400 in Hermes Agent After Switching to Claude — Third-Party Apps Now Draw From Extra Usage

Switching Hermes Agent from OpenAI to Claude returned `HTTP 400: Third-party apps now draw from your extra usage, not your plan limits.` The error text says third-party apps consume extra usage credits, while Anthropic's help article says that switch is paused, so the message and the documented state disagree. Here is how I read the error and what I did about it.

Troubleshooting Hermes Agent When It Runs Out of Context — Discard the Half-Done Work and Restart Clean

You're having Hermes Agent create a PR and it stalls with 'Context length exceeded. Cannot compress further.', and /compress does nothing. As a Hermes Agent troubleshooting tip, this post shows how to discard the half-done changes and restart clean from your committed progress to finish the PR and close the issue.

A Quick Setup for Running hermes agent Locally — macOS, Ollama, and qwen3.6:35b for a Fully Local Stack

The shortest path to installing NousResearch's hermes agent on macOS and wiring it to qwen3.6:35b running locally under Ollama. No cloud API anywhere. It comes down to three steps: pull the model, install and run setup, and launch the TUI.

Reading Hermes Agent v2026.9.7 Computer Use Internals — Capture as Target Binding and Background-First Verify → Escalate

I wrote two Hermes Computer Use posts recently (a selection review and a macOS setup log), and while watching the agent drive Calculator in the background I kept wanting to know what a single computer_use call actually does under the hood. This article is that follow-up: a code-reading tour pinned to Git tag v2026.9.7, tracing one model tool call through to a real OS action and back. Along the way it hits the five design forks I found most instructive: capture doubling as sticky-target binding, background-first with verify → escalate, mutation-safe behavior on MCP timeout, the three image return routes, and the two separate approval layers between Hermes and cua-driver.

Notes on Enabling Hermes Agent Computer Use on macOS — Seven Snags Around /Applications and TCC Permissions

As a follow-up to the selection notes I wrote earlier, I actually enabled the Hermes Agent v0.21.1 computer_use toolset on a real macOS Tahoe (26.6) machine. The docs make it look like a single hermes computer-use install command, but the cua-driver macOS installer hard-codes .app placement in /Applications, the Mac is a Jamf-managed device where I started without admin rights, and TCC binds permissions to the driver's own bundle identity. Those pieces combined to trip me up seven times. This article writes each snag down with symptoms, causes, actual log output, and the resolution, so the next person setting this up has something to grep for.

Hand It an Issue, Get Back a PR — A Hermes Agent Skill Looping Between Codex and Claude Code

When agents write the code, the highest-value human work is neither writing nor reviewing it. It is building the foundation the loop runs on, and planning the issue. Shipping this skill made that conclusion concrete.

Reviewing a PR with Claude Code and Codex at Once — Automating Multi-Model Review as a Hermes Agent Skill

A code review from a single model inherits that model's blind spots wholesale. So have independent models read the same diff, and trust the findings where they overlap. Here is how that workflow became a single command.

Spoke at Yurutech 'Something-Like-a-Study-Session' 2026.08 in Kumamoto — Autonomous Development with Hermes Agent

Notes from speaking at Yurutech 'Something-Like-a-Study-Session' 2026.08 in Kumamoto on 'Autonomous Development with Hermes Agent.' The core message: autonomous development is not about code generation, it is about automating the development loop. Humans design the goal, constraints, and evaluation function; the agent runs the loop. I also left short impressions of the two other talks.