Why I Run Hermes With OMI + Obsidian (My Stack)

A Hermes second brain is the AI memory stack I run every single day — and here's why I switched to this setup.

I tried multiple AI memory solutions over the past year.

ChatGPT memory.

Claude projects.

Custom MCP setups.

None of them stuck.

Then I built the Hermes second brain stack — OMI + Obsidian + Hermes — and I haven't looked back.

This post is the honest "why" behind it.

🔥 Want my exact Hermes second brain stack? Inside the AI Profit Boardroom, I share the OMI privacy config, Obsidian vault structure, Hermes memory prompts, and a 2-hour Hermes course covering every workflow. Plus weekly live coaching. 2,800+ members. → Get the stack here

Why Cloud AI Memory Failed Me

Three reasons.

1 — Locked to one platform

ChatGPT memory only works in ChatGPT.

Claude memory only works in Claude.

I use both + OpenClaw + Hermes + custom agents.

Cloud memory doesn't transfer.

2 — Limited capacity

Cloud platforms cap how much they remember.

After a few weeks, things drop off.

3 — Privacy concerns

For client work + business decisions + personal goals — cloud feels wrong.

I want my AI memory on my machine.

Why Hermes Second Brain Is Different

It's local.

It's portable across agents.

It's automatic.

Three principles I care about — all met.

My Stack In Detail

Three pieces.

1. OMI — captures my screen + voice automatically.

2. Obsidian — stores memories as markdown in a local vault.

3. Hermes — reads the vault on every prompt.

Plus a few extras:

Manual notes — I add deep thoughts to Obsidian directly.
Cross-agent integration — same vault works with OpenClaw + Claude.

That's the whole setup.

The Lazy Setup Path

If you want this running fast:

5 minutes — install OMI.
5 minutes — install Obsidian + create vault.
5 minutes — connect OMI to Obsidian.
5 minutes — point Hermes at the vault path.
10 minutes — test queries that need memory.

30 minutes total.

For broader Hermes context, see How To Setup Hermes Agent.

What Changed When I Switched

Three things became different overnight.

1 — Hermes started referencing my actual projects

Before: generic responses.

After: specific recommendations based on what I'm actually working on.

2 — I stopped re-explaining myself

Before: every conversation started from scratch.

After: Hermes already knows my context.

3 — Decisions got faster

Before: I'd ask questions Hermes couldn't really answer.

After: Hermes has the full picture and gives sharper input.

Watch The Walkthrough

Here's the original OMI + Obsidian + Hermes walkthrough:

That covers the wiki side — the Karpathy-style approach to structured memory.

What I Don't Like About The Setup

Be honest.

Three downsides.

1 — OMI captures more than you'd manually note

You'll get noise.

Need to review weekly to delete junk.

2 — Privacy hygiene needed

I exclude specific apps from OMI.

You should too.

3 — Setup is fiddly the first time

OMI permissions.

Obsidian vault path.

Hermes config.

It's not one-click.

But once running, it's set-and-forget.

What Made The Setup Worth It

Three things justified the friction.

1 — Time saved daily

I used to spend 30 mins+ daily writing notes manually.

Now it's automatic.

2 — Better Hermes outputs

Quality of agent responses jumped noticeably.

3 — Cross-agent portability

Same vault works for OpenClaw, Claude Desktop, Cursor.

I'm not locked into anyone's ecosystem.

How I Use Hermes Second Brain Daily

Real workflow:

Morning — open Hermes, ask "what's my priority today?"
Throughout day — work normally, OMI captures.
Calls — Hermes can reference what was said in previous calls (when I had OMI on).
Evening — Hermes summarises the day's captured decisions.

Used to need a personal assistant for this.

Now I have Hermes.

Cross-Agent Use

The Hermes second brain works with other agents too.

For OpenClaw, give the same command + path:

"Use this file path for context and memory: [Obsidian vault path]"

Cover the OpenClaw side in OpenClaw Memory Persistence.

For Claude Desktop, paste the path into a project's instructions.

Same memories, multiple agents.

What Solo Operators Should Care About

If you're a solo operator running multiple AI tools:

Stop re-explaining yourself across platforms.
Build memory once, use everywhere.
Save hours every week on context-setting.

This is why solo operators specifically benefit most.

🚀 Want my full Hermes + AI agent stack? The AI Profit Boardroom has my full agent stack, OpenClaw 6-hour course, 2-hour Hermes course, daily training, weekly live coaching. 2,800+ members. → Join here

What Teams Should Consider

If you have a team:

Each person can run their own second brain.
Don't share OMI memories across people (privacy).
Do share manual Obsidian notes via Obsidian Sync.
Each Hermes instance reads from one vault.

This way the team has shared knowledge without privacy issues.

Why I Don't Use ChatGPT Memory Anymore

Three reasons.

1 — Locked to ChatGPT

Can't use ChatGPT memories with Claude or OpenClaw.

2 — Limited what gets saved

ChatGPT decides what's "memorable" — I want everything.

3 — Lives on OpenAI's servers

Privacy concern for client work.

For all three, local Hermes second brain wins.

What This Replaces In My Stack

Things I've stopped using since the second brain:

ChatGPT memory.
Claude projects (mostly).
Notion-based context docs.
Manual prompt templates with personal context.

Replaced by one consistent Obsidian vault that all agents can read.

Daily Reality Numbers

From my own setup:

30 minutes saved daily on context-setting.
50% better Hermes responses (qualitative — sharper, more personalised).
£0 monthly cost (vs ChatGPT Plus £20+ for memory features).
Setup time: 30 minutes one-off.

ROI is positive within the first week.

Watch Hermes In Action

Here's the broader Hermes 2026 framework — second brain is one piece:

That covers the full Hermes setup, multi-agent, profiles, dashboard.

Three Things I'd Tell Anyone Setting Up

1. Be selective with OMI.

Don't enable everything on day one.

Start with screen recording, add microphone selectively.

2. Test memory recall daily.

If Hermes isn't using the vault, refine your prompts.

3. Add manual notes for deep thinking.

Don't rely only on auto-capture.

Curated notes matter.

FAQ — Why Hermes Second Brain

Why not just use ChatGPT memory?

Locked to one platform.

I use multiple agents.

Is local second brain really better?

For portability + privacy + cost — yes.

How long until it pays back?

Within the first week of daily use.

Will OMI keep recording forever?

Until you turn it off.

You're in full control.

Can I export the vault if I switch tools?

Yes — markdown is universal.

Should I encrypt the vault?

If you handle sensitive client data — yes.

For personal use — your call.

Does this work on Windows or Linux?

Mostly yes.

OMI is Mac-first.

Obsidian + Hermes work on all OSes.