Claude Code logo
Claude Code template
docs.claude.com

Founder Weekly Review

Scores last week's intentions — including the ones you quietly abandoned — then three numbers and one decision. Not a dashboard. Never sends.

NewNew8.8 KB snapshotStarter VM
Claude Code logo

What's inside

Harness

Claude Code

Plan

Starter

vCPU

1

Memory

2 GiB

Snapshot

8.8 KB

How it works · ~3 minutes

  1. 01 · Fork

    New isolated microVM on your subdomain — creator state included.

  2. 02 · Your keys

    Log into Claude Code with your own model credentials (BYOK).

  3. 03 · Ask it to work

    Open the terminal and give it a real job. You keep what ships.

30-day money-back on your first purchase · no free trial · keys never leave the VM

About this template

"Ship the pricing page" has vanished from your plan four weeks running

Nobody noticed. Meanwhile the fifteen-metric dashboard was green somewhere and red somewhere, no decision followed, and by month two it was a ritual nobody read.

The first section isn't a metric

It's last week's intentions, scored — done, not done, or abandoned, with a running streak count.

Ship pricing page — abandoned, 4 weeks running.

That line is worth more than anything else on the page, and no dashboard surfaces it. An intention quietly dropped three weeks running is a signal about capacity or conviction.

Three numbers, not a dashboard

A review reporting fifteen metrics reports none. So: exactly three.

  1. Your constraint metric — the one thing currently limiting growth. It asks what that actually is rather than assuming revenue; for most early businesses it's distribution or activation, and defaulting to revenue is how founders optimise the number furthest from what's blocking them.
  2. Cash — runway in months and the direction it moved.
  3. The leading indicator — the thing that moves before the constraint does.

Anything else goes in an appendix below the decision, so it can't displace it.

One decision, phrased as a question

"Onboarding completion fell to 41% — do we fix the flow, or change who we let in?"

Not "consider improving onboarding". And when the data supports no decision, it says no decision needed — keep executing. That refusal is what keeps the review worth reading in month three — a review that manufactures an action item every week trains you to ignore it.

Every figure is sourced

No estimating, no carrying last week's number forward, no filling a row so the table looks complete. Missing metrics show as [MISSING: source] with the decision each one blocks — which is what actually gets them supplied next week.

What you supply

Your own Claude Code login, plus whatever you track in ~/weekly/exports/ — revenue, signups, bank balance, pipeline.

Keep past reviews in ~/weekly/reviews/. The intentions loop only closes if last week's file is there; the first run has nothing to score and says so.

No integrations. Exports are the input, which is also why nothing can be sent.

Where this stops

It drafts. It does not email the review, post it, or commit you to anything in it — next week's intentions are labelled proposals.

It won't read two moving lines as causation, won't forecast from three data points, won't tell you the week was "good" (it compares to intent and lets you judge), and has no industry benchmarks.

Verified on build: the skill, the working tree, both guards, the three-numbers rule, the abandonment tracking and the one-decision rule are present on a fresh fork. Running it on your own week is yours to do.

Inside this fork

Forking copies this template into a brand-new, fully isolated microVM on your own subdomain. Here's exactly what lands in it.

  • Claude Code agent

    The upstream harness, pre-installed — same version the creator ran.

  • Starter VM

    1 vCPU · 2 GiB RAM · 10 GiB disk.

  • Creator's /persist data

    The captured persist volume is copied byte-for-byte into your fork.

  • BYOK — your keys, your VM

    Add your model API keys after forking; they live only inside your microVM.

What this agent can do

1 skill

  • founder-weekly-review

    Draft a founder's weekly review from supplied exports — three numbers, last week's intentions scored, and ONE decision to make. Never sends…

What you'll configure after forking

Secrets are scrubbed from shared templates — these are the names you supply in your agent's terminal once it boots.

Environment variables

  • ANTHROPIC_API_KEY

Your turn

Your own Founder Weekly Review, live in about 3 minutes.

Forking copies this Claude Codeagent into a brand-new, fully isolated microVM on your own subdomain — the creator's /persist state and all. Add your own keys after it boots; they never leave the box. Don't love it? Your first jurniti purchase comes with 30 days to get every cent back.

New paid VM · BYOK · 30-day money-back on your first purchase · ~3 min to provision

Starter · fork

$25/ mo

Needs 1 of your own API key