Escoffier Labs presents

Le Répertoire

How one engineer runs a 24/7 multi-agent AI stack across bare metal. Opinionated. Dogfooded. Broken-and-fixed in production.

Solomon's Cookbook: 61 production-tested recipes for running long-lived AI agents in tandem with daily coding harnesses. Not a framework. A record of what is actually deployed, what broke, and how it was fixed.

Browse the recipes Read it on GitHub
The plated edition

The 2026 Edition

The same free cookbook, edited and typeset as a book for offline reading. The diagrams and templates remain available in the public repository.

  • A proper book file Cover, contents, chapter openers, running headers, page numbers, and code that is laid out for print.
  • Full-page diagrams The stack, memory ownership, handoffs, automation, and security tiers in the Escoffier Labs print palette.
  • A bounded update window Every 2026 Edition revision published through June 30, 2027 is included with the purchase.
$39 one-time2026 Edition
See the edition

The recipes stay free here. The purchase is for the edited, designed artifact.

Bare-metal fleet, one control host A primary bare-metal agent host containing a gateway and local APIs, connected to OpenClaw nodes, a hypervisor running VMs and containers, a NAS for backups, and network mounts. BARE-METAL FLEET, ONE CONTROL HOST PRIMARY Agent host gateway loopback local APIs index + search HOMELAB Hypervisor VMs · containers · nodes STORAGE NAS scheduled backups SHARES Fleet mounts OpenClaw nodes · shares backup Solid links are live traffic. Dashed links are scheduled backup flows.
Two full-page plates from the book
Tonight's specials
II · Automation

Cron Patterns

The three-layer cron stack I actually run: systemd timers for OS plumbing, OpenClaw cron for single-shot agent tasks, n8n schedule triggers for multi-step workflows. Pick the wrong layer and you'll…

 I · AI agent stack

Multi-Model Orchestration

How to run multiple AI models in one OpenClaw setup, assign each to the right task tier, and stop burning expensive tokens on work that doesn't need them.

Updated 2026-06-05 IV · Security

Security Hardening: Linux Host for OpenClaw

Practical hardening runbook for an Ubuntu 24.04 machine running OpenClaw as an always-on AI agent. This covers firewall configuration, SSH lockdown, fail2ban, and service binding to reduce attack…

Updated 2026-04-19 III · Self-hosted infrastructure

Backup & Recovery

How to protect your OpenClaw workspace, configuration, and memory from data loss. Encrypted backups, restore testing, and disaster recovery planning.

Updated 2026-05-31
Recently updated
V · Knowledge management

The MiseLedger Evidence Pipeline: Searchable Work History From Session Logs

Agent sessions are useful until you need to find a decision from six weeks ago. MiseLedger imports local harness logs into one SQLite archive, indexes them with FTS5, and returns citable evidence…

Updated 2026-07-13 VIII · Publishing

Publish-Time Scrubbing

Scrub at the moment artifacts leave the private workspace, not after they are already public.

Updated 2026-07-13 IV · Security

Secret Management

Secrets belong in narrow local stores with boring permissions, not in prompts, repos, screenshots, or memory.

Updated 2026-07-13 V · Knowledge management

Claude Code and Codex Memory Handoffs: A Sync Path Into OpenClaw

If you run Claude Code or Codex locally alongside an OpenClaw gateway, you end up with multiple session memories. This guide describes the shared handoff format and ingester that keep OpenClaw as the…

Updated 2026-07-09
Chapters
Chapter I

AI agent stack

Multi-model orchestration. One orchestrator, many models, escalation lanes. Most of these guides assume OpenClaw as the agent framework, but the patterns generalize.

15 recipes Chapter II

Automation

Anything scheduled, hooked, or sandbox-wrapped. The decision tree for "where does this go" lives in cron-patterns.md - start there.

8 recipes Chapter III

Self-hosted infrastructure

The homelab. Hypervisor decisions, container layout, network storage, off-host backup, integration with daily-driver desktop, and surviving upgrades.

9 recipes Chapter IV

Security

Defense in depth across host, agents, network, and outbound boundary. Plus what to do when the agent does something it shouldn't.

7 recipes Chapter V

Knowledge management

How durable knowledge flows: cards, indexes, handoffs, sync. The goal is "future-me can pick up cold."

7 recipes Chapter VI

Hardware & host

The physical layer. Choosing the box, partitioning the disk, deciding what the host OS owns vs what gets virtualized.

3 recipes Chapter VII

Tools

Index of tools shipped from this stack - MCPs, dashboards, helpers - and how they fit together.

6 recipes Chapter VIII

Publishing

Reserved for docs about sanitizing and validating artifacts before they leave the host.

2 recipes Chapter IX

Philosophy

The opinions. Why this stack is shaped the way it is. What I will not do. Why this body of work is a cookbook and not a tool.

4 recipes