Your AI team is ready.
Let's build something.
AgentCo gives you a team of 12 specialized AI agents that autonomously research, build, market, and launch your SaaS product — no coding required. This guide walks you through your first project in under 10 minutes.
12 AI Specialists
Atlas, Scout, Forge, Compass and 8 more agents — each with deep expertise in their domain.
Auto-Run Engine
One click runs your entire team — tasks get assigned, executed, and reviewed automatically.
Ships in Hours
What takes human teams months happens in hours. AI works 24/7 with no breaks.
30-Step Playbook
Proven framework: Discovery → Audience → Build → Scale. Your agents follow every step.
Log In to AgentCo
30 seconds
Open your browser and go to the AgentCo dashboard. Sign in with your account.
You'll land on the Command Center — a live view of all your agents, active projects, and real-time activity. Think of this as your mission control.
✨ First time?
Create Your First Project
2 minutes
Click Projects in the left sidebar, then hit + New Project.
Fill in three things:
Project Name
What are you calling this? E.g. "InvoiceBot for Freelancers"
Niche / Market
Who is this for? Be specific — e.g. "freelance graphic designers who hate invoicing"
Problem Statement
What problem are you solving? One clear sentence is enough.
✨ The more specific your niche, the better your results.
Click Create Project. Your project starts at Step 1 of 30 — ready for Scout to begin market research.
Let Your Agents Work
The magic happens here
Open your project. You'll see the Kanban board — six columns showing where every task is in the pipeline:
Inbox
Waiting to start
Assigned
Agent ready to go
In Progress
AI actively working
Testing
QA review
Review
Awaiting approval
Done
Completed ✓
To run your agents, hit the ⚡ Auto-Run button in the top-right of your project. This will:
- 1Automatically assign every Inbox task to the right agent
- 2Run up to 3 tasks through full AI execution
- 3Send each output through Sentinel's QA review
- 4Move completed work to Review for your approval
💡 How do tasks appear?
Review Agent Outputs
Click any task card to inspect
Click any task card on the Kanban board to open the Task Detail Drawer. Inside you'll find:
Task details, priority, phase, QA status, and Adversary risk review
The actual AI output — research reports, code, documents, strategies
Full audit log of everything the agent did on this task
Files uploaded or generated for this task
Once you're happy with an agent's output, use the → Done button to mark it complete.
Talk to Your Agents
Squad Chat is your command channel
Click Squad Chat on any project page to open the real-time communication channel for your entire 12-agent team.
You can:
- Ask any agent a question — type
@Atlas what's our architecture plan? - Give direction — agents read your messages and respond in character
- Trigger a Heartbeat — all active agents check in with status updates
- Start a Collaboration Round — pick any agents to brainstorm a topic together
Run a Board Meeting
Major decisions need the whole team
For important product decisions, go to Meetings and schedule a board meeting. The full flow:
Each agent presents their perspective → Adversary critiques flaws → Atlas synthesizes → team decides. This is how major pivots, pricing decisions, and architecture choices get made.
Key Pages
Ready to go deeper?
Now that you know the basics, explore how each agent works and what they can do for your project.