Skip to content

The knowledge base for ChatGPT & Claude

Give ChatGPT & Claude long-term memory.

A simple knowledge base your AI can actually read. Save notes once. Stop re-explaining yourself.

Free to start. No credit card required.

See a few example prompts by audience

Claude
Why did we switch from REST to gRPC for the internal APIs?
hjarni: get_note
Your notes from the tech review: latency on the order service was 3x worse with JSON serialization. The team benchmarked protobuf and it cut p99 from 120ms to 40ms.
Which service migrated first?
Order service in Q1. Payments and inventory followed in Q2 once the proto schema stabilized.
Claude
What did the Series A investors say about our margins?
hjarni: search_notes
From the Sequoia meeting: they wanted gross margins above 70% before raising. You're at 64%.
Draft a one-line answer for the deck.
“Path to 72% gross by Q3 via infra consolidation and the May pricing tier rollout.”
Claude
What protocol did the Chen lab use for their cell viability assay?
hjarni: search_notes
From your literature notes: Chen et al. used a flow cytometry protocol with 48-hour incubation. 92% viability at the target concentration.
Did anyone replicate it?
Patel lab hit 89% with the same protocol last March. Your notes flagged their lower seeding density as the likely reason for the gap.
Claude
What was the name of the fisherman I interviewed in Galveston?
hjarni: search_notes
Your interview notes from March 12: Ray Guidry, 34 years shrimping the Gulf. He said the catch is half what it was in the '90s.
Pull his best quote for the lede.
“My granddaddy could fill the boat by lunch. Now I’m lucky to fill it by sundown.”
Claude
What was that ramen place near Fushimi Inari?
hjarni: search_notes
Found it. Ramen Musubi on Fukakusa street. You rated it 9/10 for the tonkotsu broth.
What time did it open?
11am weekdays, 10am weekends. Your note said to go early because the line builds fast after noon.

The problem

Sound familiar?

Claude starts every session from zero

You debugged that race condition last Tuesday. Wrote down the fix. Claude doesn't know. You re-explain the architecture. Again.

Your knowledge is scattered

Architecture decisions in Notion. Post-mortems in Google Docs. That one Slack thread with the deployment fix. Your AI can't see any of it.

Your tools do too much

Notion has 47 features you don't use. Obsidian needs plugins for everything. You just want to write something down and have your AI remember it.

More on this on the blog.

What makes Hjarni different

Different folders. Different rules.

Tell Claude and ChatGPT how to behave in each part of your knowledge base. You write the rules once. Every AI follows them.

You own the instructions.

Your bugs folder gets reproduction steps and severity tags. Your runbooks folder gets step-by-step format. Your ADRs folder gets a fixed template. Claude and ChatGPT read the rules before they touch a note.

1

Folders inherit from their parents.

A rule on a parent folder applies to everything inside it. A child folder can add more, or override.

2

Teams have their own.

Set the conventions once. Every teammate's Claude and ChatGPT follow them.

3

No prompting. No copy-paste.

The rules ride along with the note. Whichever AI is reading it gets them automatically.

Your brain

Rules stack on the way down

  • Personal brain-level

    "Prefer bullets over paragraphs. Lowercase titles."

  • Engineering

    "Use TypeScript. Link to ADRs."

  • Bugs you're here

    "Always include reproduction steps and severity."

Claude and ChatGPT read all three when they open a note in Bugs.

From a founder

Hjarni ran the whole Labeur launch. Scope decisions, landing page copy, App Store submission, launch day checklist, all in one place and all in sync with Claude. I stopped re-explaining myself to the AI. From first sketch to live in the store.
Joren De Groof, Not A Tech Company

Features

Simple on purpose.

Less software. More thinking.

Built for ChatGPT and Claude

Full REST API and native MCP server. Your AI reads and writes your notes directly. Set up ChatGPT or Claude in minutes.

Also for Cursor, Claude Code, Windsurf, and Cline.

Markdown, end to end

Notes are plain Markdown — the format your AI already speaks. No conversion. No lock-in.

Dump ideas fast

The inbox catches everything. Sort later. Don't let organizing slow down thinking.

Clip the web

One click on any article saves a clean Markdown copy to your inbox. Get the Chrome extension.

Connect your thinking

Wiki-links between notes. Build a web of knowledge, not a pile.

Share a folder publicly

Turn any folder into a read-only public page. No sign-up for whoever you send the link to. See how.

Your data is yours

Export everything as Markdown in a ZIP. Always.

Automate with Zapier

Connect Hjarni to 9,000+ apps. No code. Notes arrive automatically, and your AI can use them right away. Set up Zapier

The loop

Every new chat starts from zero. Yours doesn't have to.

Debug in Claude. Save to Hjarni. Ship the fix. Repeat.

Start the loop

Teams

One person writes it down. Everyone's AI knows it.

Share a folder. Everyone's AI catches up instantly.

A shared Hjarni folder with notes written by four different teammates

Write once, the whole team remembers

A teammate documents the deployment fix. Now everyone's AI knows it. No more "did you see that doc?"

Set rules once, the whole team follows

New hires don't start from zero. Their AI already knows your stack and conventions. No more week-one ramp.

Simple pricing

No trials. No hidden fees. Start free and stay free as long as you want.

Free

£0

forever

  • Up to 25 notes
  • Folders, tags & search
  • Full MCP & API access
Get started for free
Most popular

Pro

£8/month

Billed monthly

or £79/year (2 months free)

  • Unlimited notes
  • File attachments
  • Share folders with 2 collaborators
  • Share folders publicly
  • Everything in Free
Go Pro

Teams

£10/seat/month

Billed monthly

or £99/seat/year (2 months free)

  • Shared knowledge base
  • Team MCP & API access
  • Member management
  • Billed separately from personal
  • Everything in Pro
Start your team

Need SSO, SCIM, role-based access, or a per-team audit log? Read about Enterprise or talk to us.

Common questions

Is this just another note-taking app?

No. It's a knowledge base your AI can actually use. Your notes become live context for every AI conversation.

Why not Notion or Obsidian?

Great tools. Different goals. Hjarni is built from the ground up for AI. Custom instructions per folder. Team-wide AI conventions. An MCP server designed around your knowledge — not bolted onto a workspace. See how Hjarni compares to Notion, Obsidian, ChatGPT Projects, Claude Projects, and Apple Notes.

What's MCP?

Model Context Protocol. An open standard that lets AI assistants connect to external tools and data. Hjarni has a built-in MCP server. Your AI reads and writes your notes directly.

Which AI assistants work with Hjarni?

ChatGPT and Claude connect via MCP today. Any AI that supports MCP will work automatically. Others can use the REST API.

How do I give my AI a memory?

Add Hjarni as an MCP server in ChatGPT or Claude. Takes two minutes. Every conversation picks up where the last one left off.

What happens to my data?

Your notes live on our servers. Export everything as Markdown anytime. We don't train AI on your data. We don't sell it. It's yours.

Can I collaborate with others?

Yes. Create a team, invite members, share folders. Everyone's AI benefits from the shared knowledge.

Can I self-host Hjarni?

Not currently. It's a hosted service. That keeps things simple — for you and for us.

Write once. You both remember.

Free to start. No credit card required.

Give your AI a memory

Works with Claude and ChatGPT today. Gemini coming soon.