The Agentic AI
Handbook
An open-source guide to the architecture, patterns, and philosophy of autonomous AI agents. From OpenClaw to FlowWink — how to build agents that think, act, and evolve.
The Handbook
12 chapters covering the complete architecture of agentic AI — from theory to production.
Foreword
Why this handbook exists, who it's for, and what a single developer building in his spare time taught the world about autonomous agents.
What Is Agentic AI?
The fundamental shift from software-as-a-tool to software-as-an-agent. Understanding agency, persistence, and adaptation.
The Evolution: From Prompt-Response to Autonomous Agents
How AI evolved through five eras — from simple chatbots to self-evolving business agents.
We Run a Claw
How the Clawable project uses its own OpenClaw fork — the symbiosis between OpenClaw and Flowwink in practice.
The Claw Ecosystem
One month after OpenClaw went viral — NemoClaw, NanoClaw, SecureClaw, and what the community is building next.
From OpenClaw to Flowwink
The OpenClaw reference model — how it actually works — and how Flowwink adapted it for multi-tenant business operations.
The Agentic Control Plane
Claude Code, Cursor, Cline, Roo, Windsurf, Copilot — what they actually are, how they work, and what the thin wrapper problem reveals about moats in the AI era.
The 10 Laws of Agentic Architecture
The constraints that emerged from building production agentic systems. Skills, memory, safety, heartbeat, and self-evolution.
The Heartbeat Protocol
The 7-step autonomous loop — self-heal, propose, plan, advance, automate, reflect, remember.
The Skills Ecosystem
Skills as knowledge containers — how agents learn, evolve, and share capabilities.
Memory Architecture
The 4-tier memory system — session, working, long-term, and semantic memory powered by pgvector.
Feedback Loops
Growth loops, reflection, and self-healing — how agents compound their capabilities over time.
Stagnation and Drift
The two long-term failure modes of autonomous agents — why they stop improving and why they change in ways you didn't intend.
Human-in-the-Loop
The decision framework for when agents should act autonomously vs. when humans should approve.
Agent Governance
Who is responsible when an agent makes a mistake? How do organizations structure accountability, manage personality development, and build the new way of working?
The Digital Employee
How companies should think about AI agents — hiring, management, ROI, and organizational impact.
Agent-to-Agent Communication
How agents talk to each other — A2A protocol, authentication, discovery, and symbiosis.
ClawStack — From Theory to Swarm
How to spin up a swarm of autonomous agents on your own infrastructure in an afternoon, and connect Paperclip as the CEO that delegates to all of them.
The Future of Agentic AI
Where agentic AI is heading in April 2026 — the workforce disruption, the ethical questions, and three horizons for builders.
Partners
The organizations that made Clawable possible — and why their work matters for the future of autonomous agents.
Closing Words
What we built, what we learned, and why the people who understand this technology will shape what comes next.
Appendix B: Kilo Code
The agentic coding tool used to write this handbook — how it works under the hood, hidden gems, and why it matters in the broader ecosystem.
Appendix A: The API Layer
The three diverging inference APIs, why agentic tools enforce strict formats, and how proxies like LiteLLM preserve your freedom to switch.
The Architecture
Every successful agentic system converges on this four-layer stack.
Surfaces
I/O LayerThin wrappers that handle input/output — how the agent communicates.
Reasoning Core
CognitionThe thinking engine — prompt compiler, ReAct loop, tool router, budget manager.
Capability Layer
Skills & MemoryWhat the agent can do and what it knows — skills, memory, objectives, workflows.
Infrastructure
FoundationHow the agent runs — database, auth, storage, AI providers, edge functions.
Source Material & Partners
This handbook draws from production systems in operation today — and is made possible by two partners who live the problem we describe.
OpenClaw
Personal AI Assistant · 346k+ stars
The reference architecture for autonomous agents. A personal AI that runs on your devices, talks on 20+ channels, and evolves over time.
Flowwink
Autonomous Business Operator · MIT
The autonomous agentic CMS with FlowPilot — 73 skills, self-healing, A2A delegation, and a 7-step heartbeat protocol. The B2B reference implementation.
ClawStack
Multi-tenant OpenClaw hosting · MIT
Spin up a swarm of OpenClaw instances on your own server in an afternoon. One form, automatic HTTPS, role presets, A2A peering. From zero to agentic organization without DevOps.
Paperclip
Agent orchestration · Bundled with ClawStack
The CEO layer above ClawStack. Delegates tasks to individual Claw instances, manages the agent org chart, and runs the onboarding flow when a new Claw joins the organization.
Partners
LiteIT
liteit.se · Digital leadership & ClawStack
Swedish leadership consultancy and the company behind ClawStack. Helps organizations implement autonomous agents as digital employees — from strategy through production deployment.
Autoversio
autoversio.ai · Private AI on-premise
Uppsala-based. Builds private AI ecosystems entirely within your organization — no data leaves your walls. For regulated industries where cloud AI is not an option.