Skip to content

Karpathy brain template

This page contains a ready-to-paste template for setting up an opinionated “Karpathy brain” workspace inside Orimora. It complements the conceptual setup in the marketing docs and is meant to be copied as the README of your Brain collection.

The quick way: open /setup/structure in your workspace (or Start with a structure in the setup list on the home page) and pick As memory for my AI. Orimora creates the collections, the README and the hub documents and pins Inbox and Journal for you. The steps below do the same by hand.

  1. In Orimora, create three collections: Brain, Inbox and Journal (see the structure below).
  2. Inside Brain create a document titled README (or _index).
  3. Copy the markdown block below into that document.
  4. Adapt the project / people names to yours; keep the structure.
  5. Reference the document in your AI assistant’s system prompt with: “Before any brain operation, read the doc titled ‘README’ in the Brain collection.”

The README acts as a mini-onboarding for the LLM at every session: it learns where things live and which tags exist for ~150–200 tokens, instead of guessing.

Orimora collections are flat: a collection cannot contain another collection. Depth comes from sub-documents, so the brain is three collections, and inside Brain one hub document per area whose entries are its sub-documents.

Collection Purpose
Brain The README plus the hub documents listed below.
Inbox Rolling capture document; thoughts you haven’t processed yet.
Journal One doc per day (Journal YYYY-MM-DD), auto-created by journal_append.

The MCP tools quick_capture and journal_append populate Inbox and Journal for you; they need these two as collections of their own. Inside Brain, create these hub documents:

Hub document Its sub-documents
Decisions One per distilled decision: the question, the answer, and the why.
People One per relevant person (coffee chats, context, follow-ups).
Projects One per running project; a project’s own notes nest below it.
Reading One per book, article or talk.

In Settings → API & Webhooks, under Brain Collections, pin Inbox as the inbox collection and Journal as the journal collection.

# Brain — README
This is my second brain. It is read and written by AI assistants
through Orimora's MCP server. The conventions below let the assistant
file things in the right place without asking.
## Layout
Three collections. Collections are flat; depth comes from sub-documents.
- Collection `Inbox` — quick captures; **never search here for answers**
- Collection `Journal` — one doc per day; what happened, what I noticed
- Collection `Brain` — this README and four hub documents. New entries
are **sub-documents** of the matching hub (`parentDocumentId`):
- `Decisions` — final calls with the _why_. Use when answers are settled.
- `People` — one sub-document per person, titled `<First Last>`
- `Projects` — one sub-document per project; its notes nest below it
- `Reading` — book / article notes, one sub-document per source
## Tag schema (controlled list — do not invent new ones without me)
- `#capture` — raw inbox item, not yet processed
- `#decision` — completed decision with rationale
- `#followup` — open action item I owe someone (or myself)
- `#question` — open question I don't have an answer for yet
- `#insight` — non-obvious learning worth remembering
- `#processed` — inbox or journal entry that has been distilled
Combinations matter: `#followup` without `#decision` = open loop.
## Naming conventions for new docs
| Type | Title format | Example |
| ------------ | -------------------------------- | ------------------------------------ |
| Meeting note | `Meeting · YYYY-MM-DD · <topic>` | `Meeting · 2026-05-14 · Q3 planning` |
| Person | `<First Last>` | `Lena Schmidt` |
| Decision | `Decision · <short question>` | `Decision · Switch DB host?` |
| Project | `Project · <name>` | `Project · Karpathy brain` |
| Reading note | `<Author> — <Title>` | `K. Beck — Tidy First` |
Always prefer a noun-shaped title over a sentence. Future-me searches
by topic, not by mood.
## How the assistant should behave
1. **Search first**: before answering anything I might already have
notes on, run `search_documents`. Be explicit: tell me what you
found and what you didn't.
2. **Capture proactively**: if I share a non-trivial fact, decision,
or open question, write it down without asking permission:
- Quick thought → `quick_capture(content)`
- Today's events → `journal_append(content)`
- Structured entry → `create_document` as a sub-document of the
matching hub document in `Brain`
3. **Append, don't overwrite**: when updating an existing doc, prefer
`append_to_document` or `update_document` with a partial change.
Never replace an entire doc unless I ask for it.
4. **Confirm before deleting**: never call `delete_document` without
stating what will be deleted and why.
5. **Use the tag schema**: never invent new tags. If a topic genuinely
doesn't fit, ask me before introducing one.
6. **Mind the audit trail**: assume I will read everything you write
later. Be terse, factual, and link to sources where possible.
## Weekly distillation (Sunday, ~15 minutes)
Run this prompt against the assistant once a week:
> Read the Inbox doc and all entries from the past 7 days in Journal.
> Group them thematically. For each group, suggest:
>
> 1. Whether it belongs under Decisions, Projects, People, or Reading.
> 2. A target document title.
> 3. A 2-sentence summary suitable for that target doc.
> Do **not** write anything yet — show me the plan first.
> After I confirm, perform the moves with `create_document` /
> `append_to_document` and tag the source entries with `#processed`.
## What does NOT belong here
- Passwords, API keys, secrets — use a password manager.
- Personally identifiable information about third parties without their consent.
- Drafts I want kept hidden — use a separate private collection
with no API-key access.
## Bootstrap log
- _2026-05-14_ — Brain initialized from
[Orimora's Karpathy brain template](https://docs.orimora.com/guides/karpathy-brain-template/).
  • Replace the bootstrap-log line with today’s date.
  • Set the API key’s restrictedMode to true if other team members share the workspace, and grant the key write access only to the Brain, Inbox and Journal collections.
  • Add a stable system prompt to your AI assistant that points at this README.

Treat the README as a living convention file. After ~2 weeks of real use:

  • Drop hub documents you never wrote into.
  • Promote tags you keep wishing for.
  • Tighten naming conventions that the assistant misinterpreted.

When you change the README, tell the assistant explicitly in the next session that the conventions changed — many models cache the system-prompt context aggressively.