Anthropic dropped three updates last week that all point at the same thing — agents that watch themselves, fix themselves, and improve over time. Here’s the full breakdown plus 5 simple agents you can actually build with Claude today.
What Just Happened
Anthropic released three new features for Managed Agents on the Claude Platform: Dreaming, Multiagent Orchestration, and Outcomes. They sound unrelated until you read them together. The whole package is one bet: agents that don’t just do what you tell them, but watch themselves, fix themselves, and improve on their own.
Most of this is shipped through the developer-facing Claude Platform, so you might not be using these features directly today. But the phrase “AI agents” is about to be everywhere, and the people who already understand the structure win the next 18 months.
Research Preview
Agents that learn while they sleep
Dreaming is a scheduled process that runs in the background. While the agent is idle, it goes back through its recent sessions and memory stores, finds patterns — recurring mistakes, workflows it converges on, preferences shared across a team — and curates that into plain-text notes and structured “playbooks” the agent can reference next time.
It is not retraining the model. Anthropic doesn’t touch model weights here. What changes is the agent’s playbook — the document the agent reads when it sits down to work. So the next morning, the agent is “smarter” because its operating notes got rewritten overnight.
Why It Matters
Most AI tools you’ve used forget you the second the chat closes. Dreaming is the first formal feature where Anthropic has built compounding memory into the agent loop — so an agent that runs every day at your company gets measurably better over weeks, instead of restarting from zero.
Public Beta
A lead agent that delegates to specialist agents in parallel
A lead agent breaks a big job into pieces and assigns each piece to a specialist agent — each with its own model, prompt, and tools. The specialists work in parallel on a shared filesystem and report back. The lead agent stitches everything together.
Practical example: instead of one agent trying to handle research + writing + design + QA in one context window, you have four specialists each handling their lane, all running at the same time. Faster, cleaner outputs, less context-window bloat.
Why It Matters
For years, the “one giant prompt” was the only way to do anything complicated with AI. Multiagent orchestration is the move from “one assistant” to “a team of specialists with a manager” — the same shape as how human work actually scales.
Public Beta
Self-correcting agents that work to a quality rubric
You write a rubric describing what success looks like. The agent works toward it. A separate grader (in its own context window) evaluates the output against your criteria. If the output doesn’t meet the rubric, the agent keeps working — revising, retrying, refining — until it does.
In practical terms: instead of asking an agent to “write the email” and accepting whatever comes back, you tell it “write an email that hits these 6 criteria” and the agent self-corrects until the grader signs off.
Why It Matters
Most AI output today is “first draft and stop.” Outcomes turns the agent into something closer to a junior teammate that doesn’t hand you sloppy work — because there’s a separate evaluator inside the loop holding it to the bar you set.
All three of these features ship through Anthropic’s developer surface (the Claude Platform / Console / API). You probably won’t flip them on personally this month. But the direction is clear:
• Memory is becoming structural. Agents that remember across sessions are the new default, not a hack.
• Multi-step work is becoming a team move. One-prompt-do-everything is going away.
• Self-correction is becoming the bar. “First draft and stop” output is going to feel embarrassing within 12 months.
Translation: the agents you’ll be using inside ChatGPT, Claude.ai, and every productivity app you touch in 2026 are about to get noticeably better. The people building with Claude now are the ones who’ll know how to operate them when they show up in mainstream tools next quarter.
You don’t need the developer features to start. Every one of these can be built inside Claude Code as a Project with a context file, a workflow prompt, and a memory directory. None require an engineering background.
Agent 1
Inbox Triage
Connects to your Gmail. Every morning, surfaces the 5 emails that actually need a response, drafts a reply for each, and ignores the rest. Memory: which senders matter most, which threads you’re tracking. Saves 30 min a day for most people.
Agent 2
Weekly Performance Reporter
Pulls numbers from one platform you care about (Shopify, Stripe, Klaviyo, GA4 — pick one). Builds the same weekly report every Monday morning so you’re never opening dashboards again.
Agent 3
Personal Brief
Reads your calendar + your priorities doc + your last 7 days of work and tells you what to focus on today. Memory: what you committed to last week, what slipped, what’s urgent. The closest thing to a chief of staff you can build in a weekend.
Agent 4
Content Idea Engine
Scrapes the accounts you respect, analyzes what’s working on the platforms you post on, and generates 5 specific content ideas tailored to your niche every week. Memory: what you’ve already posted, what flopped, what flew.
Agent 5
Customer Voice Miner
If you sell something: reads your reviews, support tickets, and comments. Surfaces themes — what people love, what they complain about, what language they use. The cheapest market research you’ll ever do.
The Pattern
Every one of these agents follows the same 4-part structure I use on my own team: Context (a file that teaches the agent your situation), Connections (the data it pulls from), Workflows (the exact thing it does on a schedule), and Memory (what it saves so it gets smarter each session). Pick one of these five, build it this weekend, and you’ll have your first real agent on the team.
The Real Shift
For years, “AI agent” was a future word. Dreaming, multiagent orchestration, and outcomes are the moment it stops being future. Agents that remember, agents that delegate, agents that self-correct — that’s the loop. Whether you build the simple version this weekend or wait until it’s baked into every tool you use, the curve doesn’t reverse from here.
For Your Job
If you’re ready to set up Claude for your specific job — with custom skills, connectors, and automations built around the work you do every day — I built a bootcamp just for you.
Start the Weekend Bootcamp →Go Even Further
If you’re looking to go even further, join mine and my husband’s community group where we give you all the AI agents and systems running our businesses.
Join the Community →© 2026 Mariah Brunner. All rights reserved.