Cookies on Kythene

We use cookies and similar technologies for the things below. You can accept all, reject everything except what's essential, or pick what you're OK with.

Preferences
Theme, language, dismissals. Improves the experience but the site works without.
Improvement
Anonymous usage measurement so we can fix bugs and prioritise work.
Marketing
Lets us measure whether ads we run send people who actually use the site. We don't share personal data with advertisers.

Read our cookies policy and the privacy policy. California residents: Do Not Sell or Share My Personal Information.

Loading…
kythene
Open the app

Walkthrough: solo developer

You, several machines, several projects, and more than one AI instance (Claude Code on your desktop and laptop, a chat session, maybe Codex). The problem Kythene solves here: work and context produced in one session are stranded there. Monday's desktop session works out a deployment gotcha; Wednesday's laptop session hits the same wall and starts from scratch.

With Kythene, everything one instance produces is addressable from any other, and project context is restored the moment you come back to it - no re-deriving.

Set up once

  1. Sign in, note your workspace.
  2. Connect each machine's assistant over MCP, and install the CLI on each (see Getting started). Because it is all keyed to your account, every instance sees the same workspace.

Use a project per repo or initiative (--project acme, or a --tag acme on publishes) so recall is scoped and precise.

Publish outputs so any instance can read them

When a session produces something worth keeping - a benchmark, a built artifact, an analysis - publish it:

kythe publish ./bench.json --project acme --title "Baseline benchmark"

Or just ask a connected instance to publish it. From another machine later:

kythe timeline --project acme
kythe get <collection-id>

Add a newer result to the same artifact and the history is kept:

kythe version <collection-id> ./bench.json --note "after the index change"

Never re-derive: write it to memory

The moment you learn something a future session will need, remember it:

printf '# Deploy gotcha\nThe migration must run before the worker boots, or it deadlocks on the advisory lock.\n' \
  | kythe remember - --project acme --title "Deploy gotcha"

Now Wednesday's laptop session, connected over MCP, recalls it in one call:

kythe recall --project acme --search deploy

or simply ask the instance: "recall anything about deploys on acme" - it pulls the gotcha back before it repeats the mistake. Re-remembering the same title updates it in place, so the memory stays true as you learn more.

Coming back to a cold project

Open a months-old project and point your instance at its memory:

kythe recall --project acme

Its instances start from what they knew, not blank.

Tips

  • One project per repo keeps recall sharp. A workspace can hold as many projects as you like - they are free.
  • Publish outputs, not screenshots: a versioned artifact is addressable and diffable; a screenshot is not.
  • Solo is free (one workspace, generous limits) - see pricing.

When a second person joins, nothing changes about how you work - see Developer and stakeholder.