Wiki
Agent Teams Playbook
What builders said about running teams of AI agents: first jobs, coordinators, routines, cost and control.
(30 pages · 10 videos · updated )
What it covers
- Hubs (14)
- Syntheses (8)
- Frameworks (3)
- Blueprints (5)
Start here
Hub
All episodes
Every long-form video from Agent Teams Playbook processed into this wiki, newest first.
Hub
Permissions and human control
Hub for permissions and human control: 104 ideas from 9 episodes, organised into pages.
Every page, by type
Hubs (14)
Hub
All episodes
Every long-form video from Agent Teams Playbook processed into this wiki, newest first.
Hub
Cost and infrastructure
Hub for cost and infrastructure: 94 ideas from 8 episodes, organised into pages.
Hub
Guest directory
Everyone named on the channel as a guest, with links to their episodes and profile pages.
Hub
Permissions and human control
Hub for permissions and human control: 104 ideas from 9 episodes, organised into pages.
Hub
Questions people ask
The questions the presenters ask again and again, and where each set of answers lives.
Hub
Rolling out to a team or company
Hub for rolling out to a team or company: 113 ideas from 8 episodes, organised into pages.
Hub
Running agents day to day
Hub for running agents day to day: 80 ideas from 9 episodes, organised into pages.
Hub
Starting with agents
Hub for starting with agents: 99 ideas from 10 episodes, organised into pages.
Hub
Use cases and business models
Hub for use cases and business models: 164 ideas from 10 episodes, organised into pages.
Syntheses (8)
Synthesis
Agents and people in shared chats
Group chats and team channels where several agents and humans work in the same thread.
Synthesis
Coordinator and specialists
A chief-of-staff, chief or orchestrator agent that holds the mission, routes requests to specialist agents and reports to the human, who mostly talks to the coordinator.
Synthesis
Human in the loop
Points in an agent workflow where a person chooses, approves or reviews, and the practice of removing that oversight gradually once the flow is proven.
Synthesis
Making a business agent-ready
Letting other people's agents read, use and buy from your business, and deciding deliberately whether to block them.
Synthesis
Maturity and autonomy levels
Stages that people and organisations pass through, from chat use to agent teams that run themselves.
Synthesis
People, jobs and roles
How agents change what people do, who is hired, and which work stays with humans.
Synthesis
Selling agent services
Earning money by deploying agents for other businesses: agencies, forward-deployed engineers, roll-ups and niche products.
Synthesis
What drives agent cost
What drives agent spend: context bloat, always-on agents, parallel runs, long reports, and what a team of agents costs per month.
Frameworks (3)
Framework
Anatomy of an agent instruction
What a good instruction to an agent contains: a five-part version (role, goal, inputs, process, quality bar) and a short delegation version (goal, rules, spending or approval limi…
Framework
Building blocks of an agent
What an agent is made of: a model, a sandbox or computer, web search, an action space, connectors to data, skills, secrets, routines and memories, plus observability, feedback loo…
Framework
Coordinator plus specialists team
A coordinating (chief-of-staff) agent owns the mission and routes work to specialist agents, each with one role (research, sales, per-channel writers, designers, platform expert).
Blueprints (5)
Blueprint
A team of agents: specialists, a coordinator and clean handoffs
How to grow from one working agent to a small team with single-job specialists, one coordinator, explicit handoffs and human approval points; shown working in demos, not compared with alternatives.
Blueprint
Keep agents safe and under control
A short, ordered way to limit what an agent can touch, gate the few risky actions, cap spending and keep your own record of what it did, for anyone letting agents act on real accounts.
Blueprint
Roll agents out to a team or company
How to baseline, pilot, share and tune agents so a whole team uses them, for managers and admins moving beyond one person's set-up.
Blueprint
Run, measure and keep costs down
How to move an agent from a supervised trial to a scheduled routine, keep an eye on what it does, and stop usage and spend creeping up; for anyone running agents beyond a demo.
Blueprint
Your first agent: from one job to a working routine
How to choose one job, prove it by hand, brief an agent properly, give it tools safely and only then let it run on a schedule; for anyone building a first agent, evidence drawn from practitioners and vendor demos.
Its sources
The public pages draw on 10 videos (YouTube). Each page lists its own, with links to the moments it cites.
Show all 10
- AI Agents Could Drive 10,000x More Web Activity | Former Twitter CEO
- Grok Bot 201
- Grok Bot for Admins
- How Grok Bot Is Going To Make Me Stupidly Rich
- How I use Grok Bot to go viral on X, LI, and IG (full workflow)
- Making $$$ with Grok Bot
- Masterclass: How FDEs make $1M/yr deploying AI agents
- Phone calls with Grok Bot
- Team Bots: Grok Bots for your whole team
- The 9 Stages of AI Agent Adoption, Explained