Hermes Agent is Nous Research's self-improving AI agent framework. It's the only open-source agent with a built-in learning loop: it creates skills from experience, improves them mid-task, searches its own past conversations, and builds a deepening model of who you are across sessions.
After any complex task (5+ tool calls) Hermes automatically drafts a reusable skill document. During later runs it patches those skills in place when they’re outdated, incomplete, or wrong. Full-text session search (FTS5) plus LLM summarisation means cross-session recall is first-class, and Honcho integration builds a dialectic model of the user over time.
What you get when you run Hermes Agent on LaunchPod.
Hermes Agent connects out of the box to these platforms. Add or remove them any time from your instance.
LaunchPod clones NousResearch/hermes-agent onto a fresh Ubuntu 24.04 VM, installs it into a uv-managed Python env, and runs the gateway under systemd behind a Cloudflare tunnel. Hermes’ persistent memory and skill library live on your VM — nothing leaves without you telling it to.
Hermes Agent is open source. Here's everything you need to dig deeper into the project.
First month free. $9/mo after that. 1,000 credits included.
