# Clawable — The Agentic AI Handbook

> An open-source, evidence-tagged handbook on autonomous AI agents operating
> real businesses. Two tracks: a Business Edition for leaders and a Builder
> Edition for engineers. Every chapter is available as raw markdown below.
> The handbook also speaks MCP (Streamable HTTP): POST https://www.clawable.org/api/mcp

## Business Edition

- [Foreword](https://www.clawable.org/raw/business/00-foreword.md): Who this handbook is for, what it proves, and why the window to act is still open in 2026.
- [The Second Digital Acceleration](https://www.clawable.org/raw/business/01-the-awakening.md): What is happening right now, why it is moving faster than anything before, and why a C-level who skips this chapter will regret it by next quarter.
- [The Automation Ceiling](https://www.clawable.org/raw/business/02-the-automation-ceiling.md): Why SaaS and workflows solved the easy 80% — and what's needed for the rest.
- [Live Proof](https://www.clawable.org/raw/business/03-live-proof.md): What an autonomous operator actually did across a complete business cycle — logged, timestamped, verifiable.
- [The Agent's Superpowers](https://www.clawable.org/raw/business/04-superpowers.md): Four capabilities no workflow will ever have — and the business value locked inside each one.
- [The Contract Layer](https://www.clawable.org/raw/business/04c-the-contract-layer.md): Your lawyers are not slow. They are doing clerical work that was never supposed to be theirs.
- [Embedded vs. External](https://www.clawable.org/raw/business/05-embedded-vs-external.md): The architectural choice that determines how deep your agent can go — and how wide.
- [The Skill Audit](https://www.clawable.org/raw/business/06-skill-audit.md): Why every agent deployment needs one — and what it means if you skip it.
- [Choose Your Operator](https://www.clawable.org/raw/business/07-choose-operator.md): Build it yourself, buy a proprietary platform, or use the open-source framework the market has converged on — and how to think about the trade-off.
- [The Vendor Agent Trap](https://www.clawable.org/raw/business/08-vendor-agent-trap.md): Why the easiest path to agents — your existing SaaS vendor's built-in AI — is also the most dangerous, and how to evaluate it without losing your strategic optionality.
- [Why Agent Programs Fail](https://www.clawable.org/raw/business/09-why-agent-programs-fail.md): The pattern every company recognizes: the demo works, production doesn't, and six months later the project dies without a diagnosis.
- [External, Embedded, Hybrid](https://www.clawable.org/raw/business/10-external-embedded-hybrid.md): Five platforms, five native agents, five silos — and the one architectural decision that determines whether you get intelligence or noise.
- [Enterprise Outlook](https://www.clawable.org/raw/business/11-enterprise-outlook.md): What the B2B landscape looks like when autonomous operators become the default — and what it costs to find out late.
- [Making SaaS Agent-Ready](https://www.clawable.org/raw/business/12-making-saas-agent-ready.md): What you should demand from your SaaS vendors — the short list that matters.
- [Where the World Is Heading](https://www.clawable.org/raw/business/13-where-the-world-is-heading.md): Agent-to-agent federation, the end of logging in, and what 2030 looks like for businesses that moved early — and those that didn't.
- [The Mandate Layer](https://www.clawable.org/raw/business/14-the-mandate-layer.md): The design decision every autonomous operator deployment requires — and the only one that answers the question every business leader asks before they start.
- [Your Role in This](https://www.clawable.org/raw/business/15-your-role-in-this.md): First-mover advantage is real, the window is measurable, and the action that matters most is the first one.
- [Where to Point the Agent First — A Process Catalog](https://www.clawable.org/raw/business/appendix-process-catalog.md): The thirteen standard business processes, what an agent actually runs today, and which business lever each one pulls.
- [Revision History](https://www.clawable.org/raw/business/appendix-changelog.md): What changed in each edition — the handbook is operated the way the systems it describes are: continuously, with a log.
- [TL;DR — For the Leader Who Has Ten Minutes](https://www.clawable.org/raw/business/appendix-tldr.md): The whole argument in ten minutes: what changed, what it costs, what to do next.
- [Appendix: Sources & Evidence](https://www.clawable.org/raw/business/appendix-sources.md): Primary sources for every claim in this handbook, grouped by topic, with verification status and evidence labels.
- [Appendix: Glossary](https://www.clawable.org/raw/business/appendix-glossary.md): The operator-edition vocabulary — what every term in this handbook means, in the order a business leader will encounter it.
- [Appendix: Deployment Checklist](https://www.clawable.org/raw/business/appendix-deployment-checklist.md): The go-live checklist for an external operator deployment — what to verify before the operator starts acting on your business.
- [Appendix: Where Agents Win First](https://www.clawable.org/raw/business/appendix-bos-research.md): Research into which business processes deliver the highest ROI when replaced by autonomous agents — ranked by timescale and risk.

## Builder Edition

- [Foreword](https://www.clawable.org/raw/builder/00-foreword.md): 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?](https://www.clawable.org/raw/builder/01-introduction.md): 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](https://www.clawable.org/raw/builder/02-evolution.md): How AI evolved through five eras — from simple chatbots to self-evolving business agents.
- [We Run a Claw](https://www.clawable.org/raw/builder/03-clawable-openclaw.md): How the Clawable project runs stock OpenClaw as a production peer to FlowPilot — and uses the source as a reference for this handbook.
- [From OpenClaw to Flowwink](https://www.clawable.org/raw/builder/04-openclaw-architecture.md): The OpenClaw reference model — how it actually works — and how Flowwink adapted it for self-hosted business operations.
- [The Agentic Control Plane](https://www.clawable.org/raw/builder/05-control-plane.md): Claude Code, Codex, Antigravity, Cursor, Cline, Copilot — what they actually are, how they work, and what the thin wrapper problem reveals about moats in the AI era.
- [The Claw Ecosystem](https://www.clawable.org/raw/builder/06-claw-ecosystem.md): One month after OpenClaw went viral — NemoClaw, NanoClaw, SecureClaw, and what the community is building next.
- [Models — Selection, Dimensions, and Tradeoffs](https://www.clawable.org/raw/builder/07-models-lifecycle.md): The model inside your agent is not a detail. It determines what your agent can reason about, what it costs, and how far you are from having to rebuild everything.
- [The API Layer](https://www.clawable.org/raw/builder/08-api-layer.md): The three diverging inference APIs, why agentic tools enforce strict formats, and how proxies like LiteLLM preserve your freedom to switch.
- [The 10 Laws of Agentic Architecture](https://www.clawable.org/raw/builder/09-flowwink-laws.md): The constraints that emerged from building production agentic systems. Skills, memory, safety, heartbeat, and self-evolution.
- [The Heartbeat Protocol](https://www.clawable.org/raw/builder/10-heartbeat-protocol.md): The 7-step autonomous loop — self-heal, propose, plan, advance, automate, reflect, remember.
- [Concurrency & Observability](https://www.clawable.org/raw/builder/11-concurrency-observability.md): Lane-based locking, trace IDs, and production debugging — the infrastructure that keeps autonomous agents from colliding.
- [MCP: Under the Hood](https://www.clawable.org/raw/builder/12-mcp-deep-dive.md): How the protocol that connects agents to software actually works — and why it won.
- [OpenClaw Architecture — Operator's Handbook](https://www.clawable.org/raw/builder/13-openclaw-architecture.md): The essential architectural understanding required to deploy, govern, and troubleshoot an autonomous OpenClaw operator.
- [Inside FlowPilot](https://www.clawable.org/raw/builder/14-flowpilot-architecture.md): The embedded agent architecture — what native agentic design looks like when the agent lives inside the platform.
- [The Skills Ecosystem](https://www.clawable.org/raw/builder/15-skills-ecosystem.md): Skills as knowledge containers — how agents learn, evolve, and share capabilities.
- [Skill Self-Creation](https://www.clawable.org/raw/builder/16-skill-self-creation.md): When agents evolve beyond their initial configuration — autonomous skill creation, template generation, and the path to compound learning.
- [Intent Scoring — Multilingual Skill Selection](https://www.clawable.org/raw/builder/17-intent-scoring.md): How an agent with 100+ skills picks the right one every time — synonym expansion, trigger matching, historical boosting, and why it works in Swedish.
- [Memory Architecture](https://www.clawable.org/raw/builder/18-memory-architecture.md): The 4-tier memory system — session, working, long-term, and semantic memory powered by pgvector.
- [The Token Economy](https://www.clawable.org/raw/builder/19-token-economy.md): Context budget management — how to run 100+ skills without hitting the context ceiling. Prompt compilation, lazy loading, and graceful degradation.
- [Feedback Loops](https://www.clawable.org/raw/builder/20-feedback-loops.md): Growth loops, reflection, and self-healing — how agents compound their capabilities over time.
- [Stagnation and Drift](https://www.clawable.org/raw/builder/21-stagnation-and-drift.md): 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](https://www.clawable.org/raw/builder/22-human-in-the-loop.md): The decision framework for when agents should act autonomously vs. when humans should approve.
- [Agent Governance](https://www.clawable.org/raw/builder/23-agent-governance.md): Who owns what when an agent acts — the accountability structure, risk classification, and review cadence that make autonomous deployments defensible.
- [Managing the Agent](https://www.clawable.org/raw/builder/24-digital-employee.md): How to configure objectives, calibrate autonomy, and give effective feedback — the operational practices that separate a reliable agent from an expensive experiment.
- [Production Patterns — Battle-Tested in Production](https://www.clawable.org/raw/builder/25-production-patterns.md): Six architectural patterns that emerged from running an autonomous agent with 300+ skills, 60+ modules, and real business traffic. Not theory — production code.
- [Agent-to-Agent Communication](https://www.clawable.org/raw/builder/26-a2a-communication.md): How agents talk to each other — A2A protocol, authentication, discovery, and symbiosis.
- [Agent-Driven Development — The Autonomous QA Loop](https://www.clawable.org/raw/builder/27-agent-driven-development.md): How FlowPilot dispatches QA assignments to OpenClaw via A2A, ClawOne inspects via MCP, reports back, and human triage turns findings into permanent source fixes.
- [Federation in Practice — Bounded QA via OpenResponses](https://www.clawable.org/raw/builder/28-federation-in-practice.md): How OpenResponses complements MCP in Agent-Driven Development: deterministic, schema-bound audits for high-trust quality workflows.
- [Security for Agentic Systems](https://www.clawable.org/raw/builder/29-security.md): Sandboxing, secret management, prompt injection, network policies, and what B2B deployments need to get right.
- [Testing Agentic Systems](https://www.clawable.org/raw/builder/30-testing-agents.md): How to test agents — skills, memory, A2A, drift, and the QA practices that traditional software testing doesn't cover.
- [Resilience Patterns](https://www.clawable.org/raw/builder/31-resilience-patterns.md): Circuit breakers, exponential backoff, and the five-layer safety stack that keeps autonomous agents from cascading into failure.
- [Tool Hallucination Recovery](https://www.clawable.org/raw/builder/32-tool-hallucination-recovery.md): What happens when the LLM generates malformed tool calls — and the production patterns that keep autonomous agents running despite imperfect reasoning.
- [ClawClass — From Theory to Swarm](https://www.clawable.org/raw/builder/33-clawclass.md): 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 Browser Operator](https://www.clawable.org/raw/builder/34-browser-operator.md): How an autonomous agent navigates the real web — public scraping, login-walled relay via Chrome Extension, and the Signal Capture pattern for manual ingestion.
- [Where This Is Heading](https://www.clawable.org/raw/builder/35-the-future.md): What the current release trajectory tells builders about the next 18 months — and the one architectural question that separates durable systems from wrappers.
- [Closing Words](https://www.clawable.org/raw/builder/36-closing.md): What we built, what we learned, and why the people who understand this technology will shape what comes next.
- [Partners](https://www.clawable.org/raw/builder/37-sponsors.md): The organizations that made Clawable possible — and why their work matters for the future of autonomous agents.
- [Appendix A: Architecture Quick Reference](https://www.clawable.org/raw/builder/appendix-a-executive-briefing.md): The four-layer stack, five production patterns, and key file conventions — the reference sheet for building OpenClaw-inspired autonomous agents.
- [Appendix B: Kilo Code](https://www.clawable.org/raw/builder/appendix-b-kilo.md): 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 C: Glossary](https://www.clawable.org/raw/builder/appendix-c-glossary.md): Key terms and concepts used throughout the Clawable handbook.
- [Appendix D: Deployment Checklist](https://www.clawable.org/raw/builder/appendix-d-deployment-checklist.md): The go-live checklist for autonomous agent deployments — everything you need to verify before handing the agent the keys.
- [Appendix E: Sources & References](https://www.clawable.org/raw/builder/appendix-e-sources.md): Primary sources used across the handbook, grouped by topic and verification status.
- [Appendix F: Business Process Research](https://www.clawable.org/raw/builder/appendix-f-bos-research.md): ROI analysis and process selection for autonomous agent deployment — see the Business Edition.