35 / 35 AI agents · Press a size to see who's on the team.
You
Executive Chief of Staff
Planning and Research · 7 / 7 AI agents
Executive Chief of Staff (In the Playbook)
Goal Planning & Strategy (In the Playbook)
Plan Stress-Test (In the Playbook)
Negotiation Prep Brief (In the Playbook)
Research Brief & Digest (In the Playbook)
Meeting Recap & Actions (In the Playbook)
Meeting Time Scheduling (Not built yet)
Sales and Clients · 7 / 7 AI agents
Corporate Pitch & Proposal (In the Playbook)
RFP Review & Response (In the Playbook)
Partnership Outreach (In the Playbook)
Lead Fit Qualification (Not built yet)
Client Welcome & Intake (Not built yet)
Scope Change Detection (Just an example)
Referral Request Drafting (Just an example)
Client Support and Billing · 5 / 5 AI agents
Customer Reply Drafting (In the Playbook)
Sheet Row Logging (In the Playbook)
Late Invoice Reminders (Not built yet)
Sheet Totals & Trends (Not built yet)
Retainer Renewal Check-In (Just an example)
Project Plan & Final Check
Project Delivery · 6 / 6 AI agents
Project Plan & Final Check (In the Playbook)
Wireframe & Flow Design (In the Playbook)
Problem Diagnosis (In the Playbook)
Automation Plan & SOP (In the Playbook)
Contractor Brief (Just an example)
Client Status Update (Just an example)
Writing and Marketing · 6 / 6 AI agents
Voice-Matched Writing (In the Playbook)
Social Post Repurposing (In the Playbook)
Weekly Newsletter Drafting (In the Playbook)
Inbox Triage & Reply (In the Playbook)
Case Study Drafting (Just an example)
Presentation Outline (Not built yet)
Quality Control · 4 / 4 AI agents
Voice & Source Check (In the Playbook)
AI Agent Instruction Check (In the Playbook)
Deliverable Check (Just an example)
Client Feedback Sorting (Not built yet)
Press a size to see who's on the team.
In the Playbook
Not built yet
Just an example
Handoff to you
You do not need 35 AI agents on day one. You need the right AI agent assigned to the right job. Inside BNEDai, I operate as a team of 36: 1 human and 35 specialized AI agents that I architected, built, and run across daily operations. This guide is a worked example of a team that size. It shows how the AI agents are structured, how they work together, and how to scale your own team from one targeted function to a full operating system. The guide's team is made up, not my 35.
It is a guide, not software. The guide is one worked example that you read as a web page. The AI Agent Team Playbook, a separate product, holds the working files for 21 of the 35 AI agents.
Those are the written instructions you paste into your AI tool. 7 aren't built yet and 7 are just examples.
For anyone curious about AI agents, and for anyone who built a first one and wants a team.
A single AI agent handles one targeted function within the boundaries you set. An AI agent team connects multiple specialists so finished work moves from one step to the next, and stops with you before anything is sent. Scale your team to match the scope of the problem:
1 to 3 AI agents
Personal help
Best for one person, such as a solo business owner or executive, who wants hours back each week. It covers high-frequency admin tasks like inbox triage, meeting prep briefs, and calendar protection.
4 to 10 AI agents
Department workflow
Best for a focused business function such as sales follow-up, podcast production, content distribution, or donor research. One lead AI agent coordinates a small bench of task specialists.
11 to 35 or more AI agents
Full operating system
Best for running multi-practice company operations. This is how I run BNEDai: 1 human directing 35 functional AI agents across leadership programs, corporate philanthropy, media, and product delivery.
Who does the work
The 35 AI Agents That Run BNEDai
Every AI agent below is dedicated to a single targeted function and works within fixed rules. When you connect them in sequence, they operate as a coordinated team. I architected, built, and run all 35 of these AI agents inside BNEDai. Start with the single function that removes your second biggest time drain, then add specialists as your workflow grows.
Executive Suite & Strategy
6 AI agents
AI agentTask description
Executive Chief of Staff AI Agent
Task description: Triages daily priorities, routes tasks to specialist AI agents, and prepares your morning executive briefing.
Inbox Triage & Reply AI Agent
Task description: Sorts incoming emails by urgency, filters out noise, and drafts replies in your voice for review.
Task description: Pulls attendee bios, past email threads, and open action items into a one-page brief before every call.
Executive Strategy & Scenario AI Agent
Task description: Stress-tests new business initiatives, maps operational risks, and outlines 90-day execution plans.
Negotiation Prep & Counteroffer AI Agent
Task description: Builds pre-negotiation strategy briefs, calculates target terms, and scripts counteroffers using the L.A.T.T.E. framework.
Revenue, Sales & Partnerships
8 AI agents
AI agentTask description
Corporate Pitch & Proposal AI Agent
Task description: Drafts custom enterprise training proposals and scope documents tailored to a prospect's stated goals.
Brand Sponsorship Outreach AI Agent
Task description: Identifies aligned brand partners and drafts customized sponsorship pitches with verified audience metrics.
RFP Evaluation & Response AI Agent
Task description: Reviews incoming requests for proposals, checks qualification fit, and compiles compliant draft responses.
Prospect & Account Research AI Agent
Task description: Compiles company initiatives, leadership priorities, and recent press before sales or partnership calls.
Sales Pipeline Follow-Up AI Agent
Task description: Tracks open proposals, flags conversations that have gone quiet, and drafts timely follow-up notes.
Offer Scope & Pricing AI Agent
Task description: Structures service tiers, deliverables, and pricing boundaries for custom client engagements.
Client Welcome & Intake AI Agent
Task description: Drafts post-signature intake forms for your review, tracks incoming project assets, and prepares shared client workspaces.
Deal Terms Checklist AI Agent
Task description: Reviews incoming vendor and sponsorship agreements against your standard business terms before legal sign-off.
Content, Podcast & Media Distribution
9 AI agents
AI agentTask description
Podcast Show Notes & Packaging AI Agent
Task description: Turns raw episode transcripts into structured show notes, timestamps, and platform descriptions for YouTube and Spotify.
Podcast Guest Prep & Intake AI Agent
Task description: Collects guest bios and headshots, generates custom interview questions, and drafts pre-recording prep guides for your review.
Brand Voice Copywriting AI Agent
Task description: Drafts long-form articles, landing page copy, and keynote summaries that match your established writing rules.
Multi-Platform Social Repurposing AI Agent
Task description: Converts one core insight or transcript into tailored posts for LinkedIn, Threads, Instagram Reels, and YouTube Shorts.
Newsletter Edition Drafting AI Agent
Task description: Assembles weekly newsletter drafts from recent podcast episodes, build notes, and leadership insights.
Video Hook & Script AI Agent
Task description: Extracts high-retention clips from long-form recordings and writes promotional scripts and captions.
Search & Answer-Engine Optimization AI Agent
Task description: Structures published articles and resource guides with clear headings, metadata, and FAQ blocks for search visibility.
Editorial Calendar & Publishing AI Agent
Task description: Tracks scheduled posts across channels and flags any campaign asset that isn't queued.
Speaker Bio & Media Kit AI Agent
Task description: Keeps speaking bios, credential sheets, and press boilerplates updated with your latest milestones.
Product, Web & Daily Operations
7 AI agents
AI agentTask description
Project Delivery & Milestone AI Agent
Task description: Breaks multi-week launches into sequenced tasks, tracks dependencies, and flags overdue deliverables.
Web Layout & UX Audit AI Agent
Task description: Reviews landing pages for visual hierarchy, copy alignment, and clear conversion paths before deployment.
Site QA & Bug Triage AI Agent
Task description: Monitors form submissions, API routes, and page deployments to catch and diagnose broken workflows early.
Customer Support & FAQ Triage AI Agent
Task description: Drafts answers to common buyer questions about digital downloads, workshops, and account access from approved templates, for your review.
Workshop Registration & Attendee AI Agent
Task description: Manages event sign-up lists and drafts pre-session setup instructions and post-workshop build guides for your review.
Weekly Metrics & Conversion Summary AI Agent
Task description: Pulls traffic, newsletter growth, and product sales into a plain-English weekly performance snapshot.
Workflow Documentation AI Agent
Task description: Turns recorded build sessions and internal processes into step-by-step operating guides and checklists.
Philanthropy, Research & Specialized Workflows
5 AI agents
AI agentTask description
Donor Outreach & Stewardship AI Agent
Task description: Drafts personalized donor updates, grant follow-ups, and campaign outreach with built-in human approval gates.
Foundation & Grant Alignment AI Agent
Task description: Screens funding opportunities against nonprofit program criteria and summarizes application deadlines.
Executive Role & Opportunity Scout AI Agent
Task description: Scans target companies for leadership openings and maps candidate experience directly to role requirements.
AI Industry & Tool Briefing AI Agent
Task description: Tracks new model releases, platform updates, and enterprise AI adoption case studies relevant to your curriculum.
Participant Survey & Impact Report AI Agent
Task description: Aggregates post-workshop survey responses and calculates program completion and build metrics for impact reports.
Pick your way in
If you're curious about AI agents
Start with what an AI agent is, and why one is enough.
An AI agent is an AI tool with one job, written instructions, and limits on what it can touch. A team is several of them, passing work to each other and to you.
The guide shows you how to pick your first one: the job you do by hand most often each week.
Build one first in a free Agent Lab session, or read the first two sections of the guide below.
Managing several AI agents is a different job from building one.
One AI agent has one job. A second one means a handoff, and by the fifth you have to decide who checks the work before it reaches you.
When I rebuilt my own team, my proofreading AI agent found credentials and numbers carried over from old files into my other AI agents' instructions.
It reads every draft against my voice and my sources, and it only reads.
The guide lays out one example: six teams, two leads, you at the top, every handoff drawn.
35 is more than most people need. Start with one AI agent on the second task that takes the most of your time, and add another only when one of three cases applies.
Read one role description
This is the Executive Chief of Staff role description, as the guide shows it.
Executive Chief of Staff
In the Playbook · Stage 1
Reads your calendar and open tasks and writes a morning brief with your top three priorities.
Hands work to the right AI agent, tracks each lead with a next-contact date, and offers three meeting times.
Command
/brief /pipeline /times
Told never to
Send, delete, accept an invite, or pay. It can read your calendar but not change it.
Passes work to
You, Voice & Source Check, Research Brief & Digest, Lead Fit Qualification
Writes for others. Voice & Source Check checks it first.
If you only build one, start with the second task that takes the most of your time.
What is inside
One made-up consulting business and its team of 35 AI agents.
The guide goes top to bottom, from the first AI agent to the full team. No real client appears.
Start here
Who the guide is for, what an AI agent is, and the words you'll see.
One or many
The count: 21 in the Playbook, 7 not built yet, 7 just examples, 35 in all. And the three cases for a second AI agent.
The example business
A small consulting business with one owner, two contractors, and four kinds of clients. Swap in your own.
The org chart
Press a size from 3 to 35 and see who is on the team. Two leads, six teams, and you at the top.
Eight choices, and why they matter
Why the Executive Chief of Staff comes first and nothing sends until you say so.
35 role descriptions
Each AI agent's job, the command you type, what it's told never to do, and who it passes work to. Every role description says whether it's in the Playbook, not built yet, or just an example.
Three days in the life
A new lead, a project that ships, a late invoice. Each day ends with your decision.
Six best practices and four tests
Each practice has a line to add to your AI agent's instructions. The tests say what to give it and what a good result looks like.
If an AI agent gets it wrong
Three checklists: before you turn one on, the first month, and when something's off.
Adding AI agents, in sequence
Eight stages, from 3 AI agents to 35. It's my recommended sequence and nobody has measured it.
Your first week
A seven-day plan, from writing down the one job you do by hand most often to changing the instructions and running it again.
The setup sequence
The Setup Sequence
The sequence matters. Onboarding an AI agent works just like onboarding a new hire. If you don't set it up in the proper sequence, it leads to unintended consequences: AI agents guessing at missing context, touching tools they shouldn't access, or sending unreviewed work.
Follow this sequence for every build:
1.Define the role and operating rules
Write plain-language instructions stating what the AI agent owns, who it reports to, and what it must never do without human approval.
2.Attach reusable skills (saved instructions)
Give the AI agent specific procedures and real examples of your past work so its output matches your standard.
3.Authorize only the connectors it needs
Connect only the specific apps needed for that single function.
4.Test manually before scheduling triggers
Run the AI agent on a few real tasks and inspect the drafts before turning on automated triggers.
The guide covers the other half: the sequence for adding more AI agents, in eight stages from 3 to 35.
Read it free
Read the first two sections free.
These are the first two sections of the guide, word for word: Start here and One or many.
Start Here: What Takes Most of Your Time?
Let's start your team of AI agents with the second task that takes the most of your time. Why the second task instead of the first? The task that takes the most of your time is usually your highest-stakes core work, which carries the most complexity and nuance. Your second largest time drain is almost always a repeatable operational bottleneck. It is big enough to give you hours back immediately, and structured enough to delegate to an AI agent today.
Start here
Who this is for
You've used ChatGPT or Claude a bit. You keep hearing people say they run a team of AI agents, and you can't tell what that means or whether you need one. This guide is for you.
What an AI agent is
An AI tool with one job, written instructions, and limits on what it can touch. A team is several of them, each with its own job, passing work to each other and to you.
How to use this guide
Go top to bottom the first time. Each team has a note on why it exists, and the eight decisions below explain the choices that matter most.
Then use the buttons to see the team at 3 AI agents, 7, and on up.
Words you'll see+×
AI agent
An AI tool with one job, written instructions, and limits on what it can touch.
Team
Several AI agents, each with its own job, passing work to each other and to you.
Command
A short name you type into your AI tool to start an AI agent, like /brief.
Lead
The AI agent in charge of a team. Here, the Executive Chief of Staff and the Project Plan & Final Check AI agents each lead three teams.
Draft and hold
The AI agent writes something and stops. It doesn't send. You decide.
Voice & Source Check
The AI agent that proofreads other AI agents' drafts against your voice, your banned words, and your sources.
Checked
The Project Plan & Final Check AI agent's mark that a draft passed the Voice & Source Check AI agent and is ready for you.
Handoff
Passing work to the next step. It can go from one AI agent to another AI agent, or from an AI agent to you.
Human in the loop
A point where a person reviews and decides before anything moves on. Here, that's you.
Playbook
The AI Agent Team Playbook has the 21 working files for the AI agents marked In the Playbook.
Not built yet and Just an example
Not built yet means I plan to build it. Just an example means it's here to show the idea.
Client
Whoever pays you. Read it as customer if that fits your business better.
RFP
A formal request a client sends to companies asking for proposals.
Retainer
An ongoing paid arrangement with a client.
SOP
A written, step-by-step procedure for a job.
Tracker
The spreadsheet or tool where you track projects, clients, and due dates.
Intake
The questions you ask a new client at the start.
One or many
How many AI agents
21 in the AI Agent Team Playbook
7 not built yet
7 just examples, to show how far a team can grow
35 in all
35 is more than most people need. Start with one AI agent. Write down the one job you do by hand most often each week. That job is your first AI agent.
If you want help writing it, the Workflow Planner AI agent (stage 5 in the Playbook) takes that job and writes the procedure and a draft of the instructions.
Add a second and third only when one of the three cases below applies.
Why not just one AI agent?
Start with one AI agent. It costs less and breaks in fewer places. Every one you add is another handoff.
A second one makes sense in three cases.
You want a second pair of eyes.An AI agent that checks its own writing tends to approve it. That's why the Voice & Source Check is separate.
You want different permissions.The AI agent that reads your inbox shouldn't also be able to send.
One set of instructions got too long.An AI agent with ten jobs is harder to test. Split the jobs.
If none of those apply, stay with one AI agent.
The rest of the guide picks up at the example business and runs through the org chart, the 35 role descriptions, and the eight stages. It isn't open for purchase yet.
No, they're separate products. The guide shows how to set up AI agents and organize them into teams, with one worked example.
The AI Agent Team Playbook is the 21 working files for the AI agents marked In the Playbook.
Where do I start with AI agents?+×
Start with the second task that takes the most of your time. The task at the very top is usually your core work and carries the most nuance, so the second one is an easier first hand-off.
Write it down, then read Your first week in the guide.
How many AI agents do I need?+×
Most people need one to start: the job you do by hand most often each week. Add a second only for a second pair of eyes, different permissions, or instructions that got too long.
The guide's 35 AI agents are a worked example of how far a team can grow, not a target.
How do I organize AI agents into a team?+×
Give each AI agent one job, group them into teams under a lead AI agent, and draw every handoff, including the ones that come to you.
The guide shows one example: six teams, two leads, and a business owner who approves what is sent.
What does human in the loop mean for AI agents?+×
Human in the loop means a person reviews and decides before anything moves on. With AI agents, that person reads each draft and chooses whether it goes out.
In the guide, that person is you: every AI agent drafts and holds, and you send.
Do I need all 35 AI agents?+×
No. One is a fine start. The rest show where a team can grow.
Do I need to code?+×
No. The instructions for each AI agent in the guide are plain English.
Which AI tool do I need?+×
You need an AI tool that accepts written instructions. I haven't tested the guide in every tool. Check yours before you build.
Agent Lab builds in Gumloop, which starts with a 14-day trial, then it's a paid plan.
Will an AI agent send something without asking me?+×
Every AI agent in the guide drafts and holds, and you send. Test yours first, because an AI agent can ignore its instructions.
What if an AI agent makes things up?+×
It can. The Research Brief & Digest AI agent marks every claim VERIFIED or UNVERIFIED, with a source and a date. The Voice & Source Check AI agent checks claims against your sources.
You still read the result before it goes anywhere.
Is it safe with client data?+×
Start by pasting in only what the AI agent needs for one job. Don't connect it to a whole inbox or drive on day one.
For anything with contracts, payments, or regulated work, talk to a qualified professional first.
Is anything free?+×
Yes. The first two sections of the guide, Start here and One or many, are free to read on this page, word for word.
They cover who the guide is for, what an AI agent is, and the three cases for adding a second one. The rest picks up at the example business.
