pyrate-llama/hermes-ui — reverse-engineered prompt
Reverse engineered prompt
Build me a polished single page web dashboard for controlling a self hosted AI agent called Hermes.
I want it to feel like a futuristic glassy command center where I can chat with the agent, see streaming replies, watch live logs, browse workspace files with previews, manage installed skills, view scheduled jobs, inspect memory, see connected tools, and monitor current tasks from one place. It should run as a lightweight local web app with a simple Python proxy server, and the main UI should live in one HTML file using React in the browser.
Please include a sidebar with recent conversations, a main chat area with attachments including images and video files, a terminal or log panel, and separate sections for dashboard stats, skills, cron jobs, tools, files, and settings. Make it usable on desktop, visually close to a modern glassmorphism app, and keep the data wired through clear API endpoints so it can connect to a Hermes Agent backend. Look up current docs online if you need to.
Want more depth? Deep Reverse