The guide · playbook

Build and sell AI agents with Claude.

Building a production-ready agent used to mean months of infrastructure work — sandboxing, state management, error handling, auth — before a single user saw anything. Claude Managed Agents handles all of that for you: you define what the agent does, Anthropic runs it on their servers, and it works autonomously for hours. That changes the business model completely. Here's the full playbook, the pricing, and a technical quickstart.

The full playbook + quickstart

🔒 Drop your email to reveal the full playbook

Get the full playbook, free.

Drop your email and the niche-to-revenue playbook, pricing, and technical quickstart unlock instantly, plus every other guide and resource in the library. One email a week, unsubscribe anytime.

What just happened

Anthropic quietly launched Claude Managed Agents — and almost nobody understood what it means for people who build AI systems. Previously, building a production-ready agent meant months of infrastructure work before a single user saw anything. Claude Managed Agents handles all of it: you define what the agent does, Anthropic runs it on their servers, the agent works autonomously for hours, and it can even spawn other agents to handle subtasks in parallel.

The business model in plain English

You are now in the business of selling outcomes, not software.

What you do	What the client pays
Build the agent (one-time setup)	$1,500 – $3,000 setup fee
Monitor & maintain it monthly	$500 – $1,000/month retainer
The agent runs 24/7 on Claude's servers	Client gets the value indefinitely

10 clients × $500/month = $5,000 MRR. Maintenance time per client: roughly one hour a month. This isn't hypothetical — companies like Rakuten, Notion, Asana, and Sentry are already using Managed Agents in production, and Rakuten deployed agents across five departments in under a week each.

Step 1: Pick your niche

The fastest path to revenue is solving one repetitive problem for one type of business. Good starting niches:

Real estate agencies → lead follow-up, property listing summaries, document processing.
E-commerce brands → customer support, returns handling, product description generation.
Coaches & consultants → onboarding automation, proposal drafting, client Q&A.
Law firms / accountants → document intake, invoice processing, client intake forms.
Marketing agencies → content repurposing, reporting, client brief generation.

Pick one. Don't try to serve everyone.

Step 2: Identify their #1 repetitive problem

Book a discovery call. Ask: "What does your team do every day that you wish you could just hand off completely?" You're looking for tasks that are repetitive (done weekly or daily), rule-based (clear inputs → clear outputs), time-consuming (currently takes hours), and high-stakes enough that they'd pay to fix it. Common answers: following up with leads, processing inbound emails, generating reports, summarising documents, drafting first-draft proposals. That task becomes your agent.

Step 3: Build the agent

Go to platform.claude.com/docs/en/managed-agents/overview. You'll define four things:

The Agent — which Claude model to use, the system prompt (what it is, what it does, its rules), and the tools it can access (web search, file operations, bash, MCP integrations).
The Environment — a cloud container with the packages your agent needs, network access rules, and any files it should have.
Sessions — one "run" of the agent working on a task. Sessions persist even if you disconnect, and you can send messages mid-session to steer it.
Success Criteria (self-evaluation) — the part most people missed. You tell the agent what success looks like and it keeps iterating until it gets there on its own.

Example success criteria: "Draft a proposal for this client. The proposal is complete when it includes a problem statement, three solution options with pricing, and a clear next step. If any of those are missing, revise and try again."

Step 4: Connect it to their tools

Claude Managed Agents supports MCP (Model Context Protocol), so your agent can plug directly into HubSpot / Salesforce, Gmail / Outlook, Notion / Asana, Slack / Teams, Google Drive, GitHub, and hundreds more. Your agent doesn't just generate text — it takes real actions inside real systems. This is what separates a chatbot from an actual employee.

Step 5: Price it and sell it

Package	Setup	Monthly	What's included
Starter	$1,500	$500	1 agent, 1 use case, email support
Growth	$2,500	$800	2 agents, integrations, monthly review call
Enterprise	$4,000+	$1,500+	Multi-agent pipeline, priority support, custom training

How to find clients:

Post about it

Document what you're building on LinkedIn or Instagram. Show the process. Show the output.

Cold outreach

Pick 20 businesses in your niche. Send a 3-sentence message: what you saw their team probably does manually, what an agent could do instead, and ask if they'd want to see it.

Offer a free pilot

Build one agent for free in exchange for a testimonial and case study. Use that to close the next ten clients.

Referrals

Happy clients in a niche talk to each other. One client in real estate leads to five more.

What the costs actually look like

Token costs run at standard Claude Platform rates (same as the API). Session runtime is $0.08 per session-hour of active agent runtime. For most business use cases a well-scoped agent runs a session in under an hour, so your infrastructure cost is almost nothing compared to what you're charging. Your margin is 90%+.

The "self-evaluating agent" unlock

Most people skipped past this. In traditional agent setups, if the agent produces a bad output it just stops — you go back, fix the prompt, re-run. With Managed Agents' outcome mode (in research preview), you define what "done" looks like and the agent evaluates its own output against that definition, retrying until it gets there. In internal testing this improved task success rates by up to 10 percentage points versus standard prompt loops, with the biggest gains on the hardest tasks. For your clients that means fewer errors, less hand-holding, and more autonomous operation over time.

Multi-agent: the scaling play

Once you're comfortable with single agents, the next level is multi-agent pipelines. One "orchestrator" agent breaks a complex task into pieces and delegates to "worker" agents running in parallel. For a marketing agency: the orchestrator says "research this brand, then produce a full content calendar"; worker 1 researches the brand and competitors; worker 2 drafts the calendar; worker 3 writes copy for each post; worker 4 formats everything into a client-ready doc. All running simultaneously — delivering in minutes what used to take a day. This feature is in research preview; request access at the links below.

Technical quickstart: your first running agent

This is the part most guides skip. Here's exactly how to go from zero to a running agent.

What you need: an Anthropic Console account (console.anthropic.com), an API key (console.anthropic.com/settings/keys), and the Anthropic SDK installed.

Step 1: Install the SDK

Python:

pip install anthropic
export ANTHROPIC_API_KEY="your-api-key-here"

npm install @anthropic-ai/sdk
export ANTHROPIC_API_KEY="your-api-key-here"

from anthropic import Anthropic
client = Anthropic()
agent = client.beta.agents.create(
    name="Lead Follow-Up Agent",
    model="claude-sonnet-4-6",
    system="You are a sales follow-up assistant for a real estate agency. When given a lead's name and enquiry details, draft a personalised follow-up email and log a summary of next steps.",
    tools=[
        {"type": "agent_toolset_20260401"},  # gives access to bash, file ops, web search
    ],
)
print(f"Agent ID: {agent.id}")  # save this — you'll reuse it for every session

environment = client.beta.environments.create(
    name="real-estate-agent-env",
    config={
        "type": "cloud",
        "networking": {"type": "unrestricted"},  # lets agent access the web
    },
)
print(f"Environment ID: {environment.id}")  # save this too

# Create the session
session = client.beta.sessions.create(
    agent=agent.id,
    environment_id=environment.id,
    title="Follow up with John Smith",
)
# Open the stream and send your task
with client.beta.sessions.events.stream(session.id) as stream:
    client.beta.sessions.events.send(
        session.id,
        events=[
            {
                "type": "user.message",
                "content": [
                    {
                        "type": "text",
                        "text": "Lead name: John Smith. Enquiry: 3-bed house in Grey Lynn under $1.2M. Came through the website yesterday. Draft a follow-up email and list next steps.",
                    }
                ],
            }
        ],
    )
    # Watch the agent work in real time
    for event in stream:
        match event.type:
            case "agent.message":
                for block in event.content:
                    print(block.text, end="")
            case "agent.tool_use":
                print(f"\n[Tool: {event.name}]")
            case "session.status_idle":
                print("\n\nDone.")
                break

I'll draft a follow-up email for John Smith's enquiry.
[Tool: write]
Hi John, thank you for reaching out about your search for a 3-bedroom home in Grey Lynn...
Next steps:
1. Send email within 24 hours
2. Schedule viewing for this weekend
3. Pull comparable listings under $1.2M
Done.

agent = client.beta.agents.create(
    name="CRM Agent",
    model="claude-sonnet-4-6",
    system="You are a CRM assistant. Update HubSpot contact records and draft follow-up emails based on the task given.",
    tools=[{"type": "agent_toolset_20260401"}],
    mcp_servers=[
        {
            "type": "url",
            "url": "https://mcp.hubspot.com",   # HubSpot
            "name": "hubspot"
        },
        {
            "type": "url",
            "url": "https://gmail.mcp.claude.com/mcp",  # Gmail
            "name": "gmail"
        }
    ]
)

session = client.beta.sessions.create(
    agent=agent.id,
    environment_id=environment.id,
    title="Process new leads",
    outcome={
        "description": "For each lead: a follow-up email is drafted, the HubSpot record is updated with a next-step note, and a summary is saved to followups.txt. The task is complete only when all three are done for every lead provided."
    }
)

The agent will keep iterating until it meets that definition of done — on its own.

Full docs & next steps

Overview: platform.claude.com/docs/en/managed-agents/overview
Quickstart: platform.claude.com/docs/en/managed-agents/quickstart
Tools reference: platform.claude.com/docs/en/managed-agents/tools
MCP connector: platform.claude.com/docs/en/managed-agents/mcp-connector
Launch post: claude.com/blog/claude-managed-agents

Your move: get API access, read the docs, pick your niche and the one task you'll automate first, build a working demo within 48 hours, and send 10 outreach messages to businesses in your niche this week.

Quick reference

Term	What it means
Agent	Model + system prompt + tools + rules
Environment	The cloud container your agent runs inside
Session	One active "job" the agent is working on
Events	Messages between your app and the agent
MCP	How the agent connects to external tools (HubSpot, Gmail, etc.)
Outcomes	Define success criteria → agent self-evaluates until it reaches them

Original guide by Artem Novitckii. More from him on Instagram, in the Skool community, and across his resource library.

Build and sell AI agents with Claude.

The full playbook + quickstart

Get the full playbook, free.

What just happened

The business model in plain English

Step 1: Pick your niche

Step 2: Identify their #1 repetitive problem

Step 3: Build the agent

Step 4: Connect it to their tools

Step 5: Price it and sell it

Post about it

Cold outreach

Offer a free pilot

Referrals

What the costs actually look like

The "self-evaluating agent" unlock

Multi-agent: the scaling play

Technical quickstart: your first running agent

Step 1: Install the SDK

Step 2: Create your agent

Step 3: Create an environment

Step 4: Start a session and send it a task

Step 5: Connect to real tools via MCP

Step 6: Define outcomes (self-evaluation mode)

Full docs & next steps

Quick reference