Resources · Playbooks

Playbooks from real builds, ready to run

Step-by-step implementation guides from our client work — real code, copy-paste prompts, and the pitfalls we already hit. Every playbook is free to read.

Browse playbooks Book a call
44
Playbooks
29
Topics
34
Under an hour

Latest playbooks

Featured

Self-Host n8n and Ship a Production AI Ops Agent
Featured
Workflow Automation Intermediate

Self-Host n8n and Ship a Production AI Ops Agent

Stand up n8n on a VPS with HTTPS, connect Slack and CRM, ship a read-first AI ops agent, then gate writes with human approval.

2–3 hours Read the playbook
Zapier MCP in 20 Minutes: Let Claude Update Your CRM Safely
Featured
AI Agents Beginner

Zapier MCP in 20 Minutes: Let Claude Update Your CRM Safely

Connect Claude to Zapier MCP, allow-list CRM find/update actions, and require an explicit APPROVE before any write.

20–40 minutes Read the playbook

All playbooks

42

How much time do you have?

Any length 42 Under 1 hour 33 1–2 hours 4 2+ hours 5
Stand Up OpenClaw for Your Business
AI Agents
1–2 hours

Stand Up OpenClaw for Your Business

Install OpenClaw on an always-on host, run the Gateway as a service, connect Slack or Telegram with allowlists and pairing, and put the Control UI behind Tailscale or Cloudflare Access.

Intermediate
Production Approvals: n8n Human-in-the-Loop Tool Gates
Approvals
45–90 minutes

Production Approvals: n8n Human-in-the-Loop Tool Gates

Require human approval before an n8n AI Agent runs write tools — Slack review, real parameter display, timeouts, and deny handling for production CRM and messaging actions.

Intermediate
Self-Host Activepieces as Your Zapier Alternative
Workflow Automation
1–2 hours

Self-Host Activepieces as Your Zapier Alternative

Install Activepieces with the official Docker Compose stack (Postgres + Redis), set a public HTTPS URL, back up encryption keys, and ship your first CRM→Slack flow.

Intermediate
Business MCP Starter Pack
AI Agents
45–90 minutes

Business MCP Starter Pack

Wire HubSpot remote MCP, Slack, Sheets, and read-only Postgres into Claude with a read-first allow-list and APPROVE-gated writes.

Intermediate
Self-Host Dify: Internal RAG Chatbot for Your Docs
AI Agents
2–3 hours

Self-Host Dify: Internal RAG Chatbot for Your Docs

Deploy Dify with official Docker Compose, index internal docs into a knowledge base, and ship a citation-minded internal RAG chatbot for your team.

Intermediate
OpenClaw + n8n over MCP: Chat Brain, Workflow Hands
AI Agents
1–2 hours

OpenClaw + n8n over MCP: Chat Brain, Workflow Hands

Connect OpenClaw (or Claude) to n8n Instance-level MCP, expose only approved workflows, keep HITL on writes, and run Slack→CRM lookups through owned chat + owned automation.

Advanced
Claude Writes, Jev Decides: LinkedIn Marketing Stack
AI Agents
2–4 hours

Claude Writes, Jev Decides: LinkedIn Marketing Stack

Stand up TypeSafe Jev + Claude Code for LinkedIn radar, lead scoring, content gates, and reply-desk triage — Claude writes, Jev decides, humans approve.

Intermediate
Master Workflow Automation with Make
Workflow Automation
1-2 hours

Master Workflow Automation with Make

The complete guide to Make automation. Build production workflows that save 20+ hours per week.

Beginner
Build an AI Customer Support Chatbot
AI Integration
2-3 hours

Build an AI Customer Support Chatbot

Learn to build a customer support chatbot that actually works. Step-by-step guide from setup to deployment.

Intermediate
The AI Security Playbook
AI Security
2 to 4 hours first pass, 90 days to mature

The AI Security Playbook

How to secure AI agents, chatbots, voice agents, and n8n, Make, or Zapier automations: a step-by-step checklist for least privilege, approvals, secrets, data rules, logging, and incident response, mapped to OWASP 2026.

Intermediate
Load more

Showing 10 of 42

Inside every playbook

Written from the build, not the brochure

Each playbook comes out of a real client implementation — so you get what we actually shipped, including the parts that went wrong first.

  1. 01

    Step-by-step implementation

    The build in the order we actually shipped it for a client.

  2. 02

    Real code

    Working snippets from production, not pseudo-code.

  3. 03

    Copy-paste prompts

    The exact prompts and configs, ready to drop in.

  4. 04

    The pitfalls we hit

    What broke the first time — so it does not break for you.

From playbook to production

Want it running on your stack?

We implement these exact playbooks for clients — tailored to your stack, your data, and your team's workflow, and handed over documented.

Book a call Calculate your ROI