---
name: codebook
description: Index a private markdown corpus with a hyperdimensional (HDC) codebook — sub-second, CPU-only semantic retrieval over your own notes, with no cloud calls and nothing leaving your machine. Use when the user wants to search, trace, or answer questions over their own notes folder.
---

> **codebook — HDC note codebook · Eigen Hitchens · free skill, v1.11.0**
> Drop this file into your Claude Code `skills/` folder and invoke it by name.
> License: MIT · https://eigenhitchens.com/tools/codebook

# HDC Note Codebook

A hyperdimensional codebook library for indexing private markdown notes.
Free download. Python, CPU-only, MIT license.

## Why

Embedding-based search is slow and memory-hungry on big personal corpora. HDC
(hyperdimensional computing) gives sub-second cleanup retrieval on a CPU with 1–5 KB
per note. We use the same library to drive every retainer client's private assistant.

## Install

```bash
git clone https://github.com/squatch-c-c/hdc-note-codebook
cd hdc-note-codebook
pip install -e .
```

## Use

```bash
codebook index ~/notes --out notes.codebook
codebook ask notes.codebook "what did I write about retainers in March?"
```

## What's in the box (six codebooks)

1. **action-graph** — verbs and outcomes (5/5 cleanup pilot score)
2. **entity** — people, products, projects
3. **topic** — conceptual buckets
4. **time** — rolling temporal index
5. **citation** — quotes, links, sources
6. **mood** — sentiment / voice fingerprint

All six pack into one ~20 MB file per 10K notes. Incremental updates supported.

## Integrations

- Claude Code / Cursor / Continue.dev via the included MCP server.

## License

MIT. Source: `github.com/squatch-c-c/hdc-note-codebook`.

## Want it deployed for your team?

If you want the codebook running on your corpus, behind a private chat UI, with eval
and ops dashboards — that's a custom engagement. We scope it on a 30-minute call.

[Book a call](https://book.eigenhitchens.com/eigen/intro-call)