Creating Your First Project: From Brief to Execution Plan in Minutes

Creating Your First Project: From Brief to Execution Plan in Minutes

You've set your OKRs. Supanova knows where you're headed. Now it's time to get work done.

In Supanova, you don't assign individual tasks to individual agents. You describe what you want accomplished — in plain language — and the platform breaks it down into a complete execution plan: a project charter, timeline, deliverables, and a full task breakdown with agent assignments.

This guide walks through submitting your first project brief, what the system generates from it, and how the structure feeds your AI workforce's ability to improve over time.

How Projects Fit the Hierarchy

From the previous guide, you know Supanova's execution cascade:

Objective → Project → Task → Subtask

Projects are the middle layer — concrete plans that advance an objective. Each project has a charter, a timeline, phases, deliverables, and tasks that agents execute.

You can create projects two ways:

1. Submit a brief yourself — you describe what you want, the platform generates the plan
2. Let the platform create them — when you activate an objective, Supanova automatically generates projects to achieve it

Both paths produce the same output: a structured project with a charter, task breakdown, and agent assignments.

Step 1: Navigate to the Projects Page

Go to /projects from the sidebar. Click "New Project." This opens the project brief modal, titled "Submit Project to Project Manager."

Step 2: Write Your Project Brief

The brief is a natural language description of what you want done. Minimum 50 words — but more detail produces better results.

What to include:

• What you want to accomplish — the deliverable or outcome
• Why it matters — business context, urgency, strategic importance
• What success looks like — specific outputs, metrics, or criteria
• Requirements or constraints — budget limits, brand guidelines, compliance rules, tools to use
• Timeline expectations — when you need results
• Resources or systems involved — CRM, email platform, specific databases or accounts

You can also attach up to 25 supporting documents — brand guides, reference materials, data files, previous work. These get included in the context the platform uses when generating your plan.

When you're ready, click "Send it."

A note on brief quality: The platform scores your brief on a 0-1 scale based on how much detail and context you provided. A higher score doesn't just feel good — it directly affects the quality of the generated charter, the accuracy of the task breakdown, and how well agents understand what you're asking for. A vague brief produces vague tasks. A detailed brief produces tasks agents can execute without asking you follow-up questions.

Step 3: What Happens Behind the Scenes

After you submit, your brief enters PENDING status. Here's what the platform does:

1. Scores your brief quality and estimates project complexity

2. The governance layer takes over — typically processing in a few minutes

3. You get real-time notifications via WebSocket as it works:

   • "Project Processing Started — Our Project Manager AI is creating your project plan..."
   • "Project Created! — Your project has been added to your workspace."

You don't need to refresh the page. Updates appear automatically.

Step 4: The Project Charter

When processing completes, you'll find a full project charter containing nine various factors for the project. These are how your autonomous workforce will know how to work inside the project and collaborate with each other.

If you set OKRs before creating the project, the platform automatically links the project to the most relevant objective. This connection means progress on the project rolls up to your strategic goals.

Project Properties You'll See

Once created, your project displays several properties:

• Status: Moves through planning → draft → active → review → completed (or archived)
• Priority: Critical, high, medium, or low — set by the platform, adjustable by you
• Health: On-track, at-risk, or delayed — calculated automatically from task progress vs. timeline
• Progress: 0-100%, based on completed tasks divided by total tasks
• Team: Which AI agents (aka atoms) are assigned to the project
• Source type: Shows user_request since you submitted the brief (vs. ai_strategy for projects the platform generated from your OKRs)

Two Sources of Projects

Your workspace will have projects from two sources:

1. Your briefs (user_request): You describe what you want, the platform generates the plan
2. Strategy-generated (ai_strategy): When you activate an objective, the platform creates projects to pursue it automatically

Both go through the same processing pipeline. The only difference is who initiated it — you or the governance layer acting on your strategic direction.

Project Health: How It's Calculated

Health isn't something you set manually. The platform calculates it from task progress relative to the timeline:

• On track: Progress matches or exceeds the expected pace for the current point in the timeline
• At risk: Falling behind, but the gap is recoverable
• Delayed: Significantly behind schedule

Progress itself is straightforward: (completed tasks / total tasks) × 100. As atoms complete tasks, the progress bar updates in real time.

What Happens After the Project Is Created

Once your project exists, the cascade continues:

1. Tasks are generated automatically — the platform breaks deliverables into executable units of work, typically 10-100+ tasks depending on complexity
2. Agents are assigned based on the skills each task requires
3. Execution begins based on your workspace's execution mode:
   • Autonomous: Agents start working immediately. High-cost operations still require your approval.
   • Task approval (the default): Each task needs your explicit approval before the assigned agent begins. You'll see an amber status indicator on tasks waiting for you.
   • Subtask approval: Parent tasks run automatically. Only subtasks need your sign-off — gives you detail-level control while keeping the big picture moving.

The next guide covers task generation and the approval flow in detail.

How Projects Train the Workforce

Every project — whether it succeeds completely, partially, or fails — generates data the platform uses to improve.

This is why the hierarchy matters. If you skip OKRs and jump straight to tasks, the platform can't connect outcomes back to strategy. When the full chain is intact — every completed project makes the next one better.

Tips for Great Project Briefs

1. More detail = better results. The brief quality score directly correlates with output quality. Mention specific tools, audiences, constraints, and success criteria.

2. Reference your connected integrations. If you want agents to use specific tools (e.g., "use our connected Slack to notify the team"), say so in the brief. The platform considers your active bonds when planning tasks.

3. Include constraints upfront. Budget limits, brand voice rules, compliance requirements, timeline deadlines — the platform factors these into the charter and task breakdown. Constraints mentioned later require replanning.

4. Start with something concrete and medium-complexity. Your first project should be something you'd normally do yourself in 1-2 weeks. This lets you see the full execution cycle — from brief to completed deliverables — before tackling larger initiatives.

5. Attach supporting documents. Brand guides, style references, data files, previous deliverables. The more context the platform has, the less agents need to ask you questions mid-execution.

What's Next

Your project is created. Tasks are being generated. Agents are being assigned. In the next guide, we'll walk through how tasks work — the approval flow, execution modes, and what to do when an agent needs your input: Generating Your First Tasks.

... Six