Skip to content
← Back to the familyEXPERIMENT / 08/v0.8.0

THE ONE THAT REMEMBERS THE WHOLE STORY

Cortexv0.8.0

New session.
Same brilliant memory.

Local-first, token-efficient persistent memory for Claude Code — progressive recall with ~66% fewer tokens.

YOUR TERMINAL, PLEASE.
uv tool install "cortex-claude[code]" --python 3.12 && cortex-claude setup

Read the docs on GitHub

Free · MIT · uv picks a compatible Python for you. pip works too.

Follow the story — release notes
context is a love language.
CortexSMALL APP. BIG IDEAS.

Because “as I was saying…” shouldn’t be a workflow.

i remember why you opened that terminal.

Made for Claude CodeLocal SQLite storageZero API calls100% CORTEX LABS ENERGY

01 / MORE THAN A PRETTY ICON

Big little
superpowers.

6 ways to make your day easier.
Yes, we got a little carried away.

SUPERPOWER / 01

Progressive recall

3-layer retrieval (facts → summaries → full content) stops at the cheapest layer that answers — facts cost 5–15 tokens.

SUPERPOWER / 02

Knowledge graph

Auto-extracted structured facts with entity normalization, deduplication, and multi-hop traversal across connections.

SUPERPOWER / 03

Code graph

Tree-sitter indexes symbols across 7 languages — ~100 tokens to look one up vs 1000+ to read the file.

And a few more tricks.

Go on. Open one.

04

Smart auto-capture

Hooks capture results from your tools and a background daemon saves them in about 0.3 seconds.

05

Self-organizing clusters

Memories group into semantic sub-graphs with human-readable labels — browse topics top-down before drilling in.

06

Decay & temporal awareness

Unused memories fade, contradicted facts lose confidence, and timestamps keep recall fresh instead of stale.

The whole story, for the curious.

A closer look at what makes Cortex tick.

Cortex gives Claude Code persistent memory through a 3-layer retrieval system that returns only what's relevant, using ~66% fewer tokens on average. It automatically extracts structured facts from your memories and code into a queryable knowledge graph, with entity normalization and multi-hop traversal. Everything stays local — SQLite + Python, zero API calls, zero cost. It's the memory the rest of the family plugs into, and the project that named the lab.

02 / LET’S MAKE THIS OFFICIAL

Are we
your type?

A few practical details.
No awkward first-date questions.

THE COMPATIBILITY CHECK

A little space.
A lot of potential.

For Claude Code.
Made to feel right at home.

SMALL LAB
BIG CARE
Works with
Claude Code, via MCP
Runs on
macOS, Linux, Windows · Python 3.11 to 3.13
Price
Free — zero API calls, zero cost
Dashboard
A web UI at localhost:37800
Your data
Stays local, in a file on your machine

YOUR NEXT FAVORITE TOOL IS RIGHT HERE.

Give your next session a head start.

One install. A lot fewer “let me explain this again” moments.

YOUR TERMINAL, PLEASE.
uv tool install "cortex-claude[code]" --python 3.12 && cortex-claude setup

Read the docs on GitHub

Free · MIT · uv picks a compatible Python for you. pip works too.