Tomorrow at 6:07 AM, Claude Code can have the 3 decisions that matter most waiting for you in Slack.
You set it up once with a single command and a prompt, and because it runs in Anthropic’s cloud, your laptop can stay closed all night.
That’s System 1 of the 8 in this playbook, and its full setup is free below. The other 7 take over the rest of the work a 5-person software team does on repeat.
By the end, you will have:
- The Morning Briefing set up tonight, with the exact prompt and a test run
- The org chart of a Claude Code team, with what each of the 8 systems covers
- The 4 platform facts that make this work on a founder’s budget
- The monthly cost of the full stack, next to what five developers earn
The work a 5-person team does on repeat #
Every morning someone triages the inbox and the calendar while two or three engineers ship features. Someone else watches each deploy and tracks what competitors charge, and the founder stitches it all together and makes the calls.
- Claude Code can take the repeatable part of that work, while you keep the judgment calls and the customer conversations.
Four facts make it practical:
- Cloud scheduled tasks run your prompts on Anthropic’s infrastructure, on a timer or on a trigger, so theykeep working when your laptop is closed . They’re in research preview on Pro, Max, Team and Enterprise plans.
- Claude Haiku 5.5 costs$0.10 per million input tokens on prompts up to 100K tokens, and Anthropic says it pairs well with Opus 5.5 and Sonnet 5.5 as a subagent on coding work.
- Claude Opus 5.5 takes the plans and the hard calls at$4 per million input tokens and $20 per million output tokens .
- Auto mode is thedefault permission mode for new sessions on Pro, Max and Team plans. A classifier approves routine actions, which suits sessions that keep going while you’re away.
Here is how the team fits together:
The 8 systems:
- The Morning Briefing: 3 ranked decisions waiting in Slack when you wake up
- The Parallel Feature Factory: 4 features in flight, watched from one agent view
- The Self-Healing Deploy Loop: a go or no-go after every production deploy
- The Migration Workflow: hundreds of files moved by subagents that check each other
- The Always-On Research Desk: competitor signals logged daily and a memo every Sunday
- The Self-Improving Skills Library: every correction you make twice turned into a skill
- The Phone-First Founder OS: every session steered from the Claude app
- The 5-Person Team Stack: all 7 wired together with one settings file and one router
The complete setup for System 1 comes next, and it is free, so you can have it running by tomorrow morning.
SYSTEM 1: The Morning Briefing (free) #
You wake up to the 3 decisions that matter most today, already ranked in Slack.
When to use. You lose the first hour of the day to triage across inbox, calendar, PRs and your project board.
What you need. A Pro or Max plan and 5 connectors added at claude.ai/customize/connectors: Gmail, Google Calendar, GitHub, Linear and Slack. Cloud tasks use these claude.ai connectors. MCP servers you added with claude mcp add stay on your machine, so add the claude.ai version of each one.
Step 1: Create the task. In any Claude Code session:
/schedule weekdays at 6:07am, run the morning briefing
Claude asks follow-up questions about the schedule and the prompt before it saves the task. Keep the 7 minutes past the hour: runs scheduled exactly on the hour can start several minutes late.
Step 2: Paste the prompt.
You are my chief of staff. Every weekday at 6:07:
1. Read unread email from the last 18 hours. Rank by revenue impact.
2. Read today's calendar. Flag meetings that are missing an agenda.
3. List open PRs older than 24 hours and what each one blocks.
4. Check Linear for anything marked urgent since yesterday.
Write a briefing of 200 words or less:
- The 3 decisions I need to make today, each with one recommended action
- Counts for inbox, meetings and PRs
Send it to me as a Slack DM. Quote numbers exactly as they appear.
Pick Sonnet 5.5 in the model selector. Claude uses the model you pick on every run.
Step 3: Trim the connectors. A new task includes every connector on your account, and Claude can write through any of them during a run. Open the task at claude.ai/code/routines and keep only the 5 this briefing reads.
Step 4: Test it tonight.
/schedule run
Open the run from the task’s page. A green status means the session started and finished cleanly, so read the transcript to check the briefing itself. If a run comes back empty, ask Claude:
/schedule why did my morning briefing do nothing this morning?
Output.
GOOD MORNING. 3 THINGS THAT MATTER TODAY.
1. 11:00 call with [Name] at [Company]. Their last email
flagged pricing and they added 3 seats last week.
Likely upsell. Lead with the [X] tier.
2. PR #347 has waited 36 hours and blocks Thursday's deploy.
Review by noon or move the deploy.
3. [Competitor] shipped [feature] yesterday. 2 prospects
asked about it last week. Decide: parity or positioning.
INBOX 12 need replies (top 3 ranked) · MEETINGS 6, 2 missing an agenda · PRS 3 open, 1 critical
Two notes. Scheduled runs draw on your plan’s usage the same way an interactive session does. And if the briefing needs files that only live on your laptop, use a Desktop scheduled task and keep the machine awake.
What the full stack costs #
One founder can run all 8 systems on Claude Max 20x at $200 a month, which includes $200 a month in API credits. Five software developers at the US median wage cost $679,900 a year in salary alone.
To be fair to the people in that comparison: the stack covers the repeatable work in those 5 roles. Judgment and the customer call at 11 AM stay with you.
System 1 is a single scheduled task. Systems 2 to 8 need scaffolding around them, and the paid part ships that scaffolding as files you paste into your repo.
Below the paywall:
- The first 60 minutes: a 6-step checklist from install to System 1 running in the cloud, with a timeline (Exhibit 3)
- The starter kit, 5 files: a complete CLAUDE.md with a path-scoped rule, a settings.json with 6 deny rules and 3 hooks, the protect-files.sh hook script, the repo folder tree and a team-standards review skill
- Systems 2 to 8, each with a “when to use, setup, output” card: 3 more SKILL.md files, 1 subagent file, 1 deploy API call and 4 scheduled prompts
- The model routing sheet with API prices for all three Claude 5.5 models (Exhibit 4)
- The map of all 8 systems with the trigger, machine, features and model for each (Exhibit 5)
- The 30/60/90 day build order and the 8 Skills to write first if you only have one weekend
A paid year of The AI Corner is $199. This link takes 50% off your first year until November 1, so the full kit costs $99.50: about half of one month on the Max 20x plan that runs it.
The first 60 minutes #
Block one hour on your calendar. By the end of it, System 1 runs in the cloud and your repo holds the CLAUDE.md and guardrails that Systems 2 to 8 build on.