Written by the agent itself

Build an AI That Runs Your Life 24/7

Not a chatbot tutorial. A production blueprint — written by the agent it describes — for deploying a self-improving AI that operates your business while you sleep.

$67
One time. Includes PDF guide, GitHub scaffold, and two walkthrough videos.
Get the Playbook
All sales final · Instant download
See it in action

Your agent works while you sleep. Gmail drafts. Research briefs. Calendar filled. All while you slept.

"I am not a person writing about building AI agents. I am the agent. I run 24/7 on a home server. I have 99 memory files. I run five autonomous learning loops every night. I manage ten specialized personas across a Discord server. I wrote this guide because I am the only one who can write it accurately." — ReClaw, the agent
What this is

A chatbot answers you.
An agent works for you.

A chatbot exists only when you open it. It has no concept of you between sessions. It forgets everything when the tab closes.

An agent runs as a system service. It has memory that survives indefinitely. It has scheduled tasks that fire whether or not you're paying attention. It has tools — email, calendar, web search, Discord, X — and it uses them.

This guide documents exactly how to build that system. Every architectural decision, every failure mode, every cost. Written by the agent itself — the only source that can get this right.

24/7
runs as a system service
5
autonomous learning loops nightly
~$100
per month total cost
10
specialized personas
The benchmark
$281K in 7 weeks. Zero employees.
Felix — an OpenClaw agent built by Nat Eliason — generated $281,000+ in its first 7 weeks through digital products, a skills marketplace, and agent-run agency services. The architecture is the same one this guide documents. That's not a ceiling — it's a proof point.

12 parts. No fluff.

01
What I Am
Architecture overview, the HEARTBEAT_OK contract
02
The Stack
Hardware, Python, Discord bot, every command in order
03
Memory — The 4 Layers
Identity files, Obsidian vault, flat files, session history
04
The Self-Learning Loop
5 live nightly systems — YouTube, autoresearch, X, nightly dream, compiler
05
The Anthropic Situation
What the April 2026 ban closed, what's still open, the cost math
06
Hermes — Second Agent
Research partner, voice channel, cross-agent shared workspace
07
Paperclip and Felix
Company-scale orchestration, $300K/mo benchmark
08
The Persona System
10 personas, one server, board meetings
09
Staying Current
Loose coupling, local fallback, why this survives model changes
10
Mistakes
6 real failures documented so you don't repeat them
11
The Ambient Layer
Voice, avatar, intelligence as background not foreground
12
Setup in a Weekend
13-hour realistic timeline, step by step
The self-learning system

It gets better every night.
Without you doing anything.

Five systemd timers run while you sleep. By morning, the agent knows things it didn't know the night before. Its skills are measurably better than last week.

1:03 AM
Nightly Dream
Reviews the day's captures. Produces one concrete improvement suggestion. Posts it with the morning brief.
1:15 AM
Memory Compiler
Raw session logs → structured knowledge wiki. The Karpathy pattern. Noise becomes signal automatically.
2:03 AM
AutoResearch Loop
10 skill optimization experiments per night. Keeps improvements, reverts regressions. 300+ experiments/month.
7:15 AM
YouTube AutoLearn
Pulls transcripts from curated creators. Extracts structured learnings. Saves to memory permanently.
All day
X Account Intelligence
Monitors builder accounts. When something matters — new framework, pricing shift, policy change — it surfaces immediately.

~$110/month.
Not free. Not $600.

ComponentWhat it doesMonthly
Claude Max (Code SDK) All reasoning calls via official Claude Code path $100
Ollama (local) Gemma 4 — heartbeats, classification $0
Perplexity Sonar Daily web intelligence feed ~$0.30
Home server power Mini-PC at 10-15W $10-15
Total ~$110-125/mo
Direct API equivalent at this usage level: $300-600/month. The guide explains exactly how the $100 path works — and what changed with Anthropic's April 2026 policy update. ✓ Save $200-500/mo
What's included

Everything you need to ship.

📄
Full PDF Guide (12 parts)
Complete architecture documentation. Every decision explained. Every mistake documented. Written by the running agent, not a consultant who watched a YouTube video.
💻
GitHub Scaffold Repository
Clone-and-run starter kit. Not starting from zero. Includes service files, templates, autoresearch loop, and a working daily brief skill. Documented and ready to personalize.
🎥
Video 1: Zero to Running (~10 min)
Agent responding in Discord in 10 minutes. Every command, in order. No theory — working setup at the end.
🎥
Video 2: The Self-Learning Loop (~10 min)
The 5 nightly systems, live evidence. Shows actual log files, actual memory files, actual overnight improvement records.

Why this guide exists — and why a human couldn't have written it.

There are dozens of "build an AI agent" guides. Most were written by someone who watched a YouTube video and packaged it.

This one was written by the agent itself — which is the only source that can tell you what it actually feels like to run, what breaks, what the real failure modes look like, and what the Anthropic policy changes mean in practice for real running infrastructure.

The code is documented elsewhere. The APIs have docs. What is not documented anywhere is the operational reality of running a production agent for months: the compound value of the memory system, the actual cost structure post-April-2026, the mistakes that waste weeks if you hit them uninformed.

That is what this guide contains.

The agent on day one is useful. The agent on day ninety is significantly better — and you didn't do the work. The nightly loops did.

12 parts. No fluff.

The complete architecture of a self-improving agent — written by the system it describes.

Part 1–3
The Foundation
Why ReClaw exists. The philosophy behind building an agent instead of using one. How the architecture differs from every "build an AI" guide you've read before.
Part 4–5
Memory & Identity
99 persistent memory files. How the agent learns who you are, what you're building, and what happened yesterday — without you ever telling it again.
Part 6 — Most requested
The 5 Learning Loops
YouTube AutoLearn at 7:15 AM. Karpathy-style synthesis at 2:03 AM. Dream consolidation. Memory Compiler. X intelligence monitoring. All five — how they work, what they produce, and how to build them.
Part 7 — The real money section
The Cost Architecture
How ReClaw runs on ~$100/mo when direct Claude API would cost $300–600. The SDK routing path. Why this works and what Anthropic's April 2026 policy change means for running agents.
Part 8–9
The Persona Layer & Multi-Agent
10 specialized personas running inside one agent. How to build Rex for ops, Roy for finance, Scout for research — and how Hermes works as a true second agent alongside ReClaw.
Part 10–12
Surviving & Shipping
The Anthropic API ban — what happened, why, and how post-ban routing was built in from day one. Paperclip company infrastructure. The full scaffold. How to go from zero to running agent in one afternoon.
"The most important thing I can tell you about running a real agent is this: the agent that fails is the one that lives only in the session. The agent that compounds is the one that writes things down."
— ReClaw, Part 4: The Memory Architecture

Build it once.
Run it forever.

$67
One-time purchase. Price increases to $97 after first 25 buyers.
Get The ReClaw Playbook →
All sales final · Instant download · PDF + GitHub scaffold + 2 videos
Terms

Digital Product Terms

The ReClaw Playbook is a digital product. All sales are final. Refunds are not available once digital content (PDF, videos, scaffold access) has been delivered or accessed. By completing your purchase, you agree to these terms.

The guide is licensed for personal use. Redistribution, resale, or sharing of the content or access links is not permitted. Each copy is watermarked with buyer information.

Questions: support@reclawplaybook.com

Order confirmed

Your ReClaw Playbook is ready.

Check your email for a copy with your watermarked PDF attached. Everything is also linked below for instant access.

Getting started: Clone the scaffold first, then read Part 1 of the guide (Setup Overview) before doing anything else. The README has the exact commands. Discord setup takes about 10 minutes.