AI agents are only as good as what you ask of them.
I'm Axel Verne. I write short, field-tested playbooks for getting real work out of the Hermes agent — no hype, no 400-page theory. Free tips below; the full systems are in the books.
Three tips you can use in the next five minutes
The same principles behind the books — sized to fit in a coffee break.
Give the agent an exit
Agents fail loudest when forced to guess. Add one line to any task: "If the answer isn't in the provided material, say so instead of guessing." Hallucination drops immediately — and you find out what's actually missing.
Front-load the constraints
Don't bury the deadline, the format, or the word count at the end. Put every constraint in the first two sentences. Models weight openings heavily — the same request moved to the front measurably improves compliance.
Make it re-runnable
A good prompt is a tool, not a message. Save your best ones with placeholders — "Summarize {{URL}} in {{N}} bullets for {{AUDIENCE}}" — and reuse them. That's the difference between chatting and operating.
The books
Both are instant downloads — PDF + EPUB, no subscription, no seat limits. Every future update lands free in the same files.
Hermes Agent for Complete Beginners
Zero to a running agent without touching a terminal. Install, first tasks, safety rails, and the ten workflows everyone should start with.
- PDF + EPUB, instant download
- Written for day one, not day one-hundred
- Free lifetime updates
58 Prompts for Power Users
The full playbook: 58 field-tested prompts organized by job — research, writing, code, automation — each with when-to-use notes and failure modes.
- PDF + EPUB + searchable Prompt Library
- Every prompt re-runnable with placeholders
- Free lifetime updates
The Complete Collection
Both books together, plus the searchable library of all 58 prompts — one purchase, every future update included.
- Everything in both books
- One download, all formats
- $8 less than buying separately
Prefer whop? The same books live at Hermes Agent Playbooks on whop — with a free prompt library post and the community alongside.
What I actually run
I don't recommend tools I don't use daily. These are the picks that survived — each explained honestly on its own page.
⚙️ Hermes Agent
The agent the books are built around — local-first, scriptable, and the reason the prompt library exists.
🛠️ The official docs
Every install command and config option, straight from the source — the same references the books point to.
📚 The community & the lab
The Nous Discord for questions the docs don't answer, and Nous Research's site for where the agent is heading next.
Affiliate disclosure: some links on this site are affiliate links — if you buy through them, I earn a commission at no extra cost to you. I only link tools I use myself, and my books never require any of them.