for Claude Code

Agent OS — Memory & Continuity for Claude Code

Your agent has amnesia, and you pay the tax every morning. Agent OS makes it write down what it learned before the session dies — and read it back the moment the next one starts.

Get Agent OS $29
Instant download · 30-day refund, no questions
Python 3.9+ · macOS, Linux, Windows

Be the first to review it.

without
$ claude
> what were we building?
I don't have context from
previous sessions. Could you
explain the project?
— every morning —
with agent os
$ claude
1. Phase: auth rewrite
2. In progress: token refresh,
   2 tests still red
3. Next: fix the expiry edge
   case, then ship
— briefed before you ask —
the structural mistake

Most people treat CLAUDE.md as a README

That is why it stops working at project three. A README describes. What you need is a constitution — and a constitution without enforcement is just a wish written down.

Agent OS splits the one file almost everyone overloads into three, divided by how fast each one changes:

Law CLAUDE.md — rules only, changed deliberately never compressed
State PROGRESS.md — phase, in flight, verified compressed after 24h
Wisdom LESSONS.md — corrections and traps, append-only never compressed

The separation is the whole trick. State churns daily, so it has to be compressed or you bleed tokens on every session start. Wisdom is permanent, so compressing it destroys the exact thing you kept it for. Share one file and no setting satisfies both — you either lose the lessons or pay for the bloat.

what you get

Files, hooks, and scripts — not prompts

Four document templates

Master CLAUDE.md with a project registry, plus the three-file continuity kit. The registry is what lets the agent answer questions about project seven while you're sitting in project two.

Two working hooks

One denies every write to .env and friends. One runs your tests the instant code changes and feeds failures straight back to the agent. Enforcement, not polite requests.

Scaffolding script

new_project.py lays down the full doc kit and hook wiring in one command. It never overwrites an existing file, so it's safe on a project that's already half set up.

Registry audit

registry_scan.py walks every project you own and reports which ones are missing docs or have gone stale. Reads only filenames and timestamps — safe to run against folders holding secrets.

The 12 Rules

Each rule written next to the specific failure that produced it. A rule without its scar tissue gets deleted by the next person who finds it inconvenient.

Scaling guide

What breaks first at 1, 5, and 20 projects — session amnesia, then cross-project blindness, then repeated mistakes, then token bleed, then silent write gaps. In that order.

verified, not asserted

Every script checks itself

You should not have to trust a stranger's hook with your repository. Run these the minute you unzip it — no install, no dependencies, standard library only:

$ python hooks/guard_secrets.py --demo
guard_secrets: all checks passed
$ python hooks/verify_on_save.py --demo
verify_on_save: all checks passed
$ python scripts/new_project.py --demo
new_project: all checks passed
$ python scripts/registry_scan.py --demo
registry_scan: all checks passed
the obvious objection

Doesn't Claude Code have Auto Memory now?

Fair question — I asked it myself two days after launch. Auto Memory and claude-mem (71k stars, free) are both real and both free. Here's what they don't cover:

They solve session memory — what happened last time. They're not opinionated about rules you want enforced, not just remembered. A PreToolUse hook that hard-blocks writes to .env is a wall. A memory note that says "don't touch .env" is a suggestion the model can still talk itself past.

They're single-project. Run more than one thing at once and the failure that actually costs time isn't forgetting — it's one project's agent silently not writing to the shared log the others read from. Neither has a concept of "other projects."

If you run one project and just want to stop re-explaining yesterday, native Auto Memory genuinely covers that — save your $29. This is for the point past that, where "remembered" isn't the problem anymore and "enforced, and consistent across projects" is.

honest fit

Worth $29 for some people, not for others

Buy it if

  • You run Claude Code on more than one project
  • You've re-explained the same context more than twice
  • You've watched the agent repeat a mistake it already made
  • You want the structure decided, not another framework to learn

Skip it if

  • One weekend project, one session — you don't have this problem yet
  • You want "act as a senior engineer" prompts. There are none here.
  • You want a plugin to install. This is markdown and four Python files.

Stop explaining your project every morning

Fifteen minutes to set up. The quickstart ends with three concrete tests, so you'll know it's live rather than hoping.

14 files · zero dependencies · 30-day refund