Sweet 16 Agent Pack

Catch the angles most bettors miss before tip-off.

Use 16 specialist prompts to test momentum, pace, shooting, injuries, coaching, and market perception before you make the call.

Preview first-round matchups

Works with ChatGPT, Claude, and Gemini. Instant delivery. No subscription. 7-day guarantee.

Inside the $9 pack

MOMomentum
PAPace & Tempo
3P3-Point Mismatch
INInjury & Availability
CSCoach & Scheme
ATBrand Perception & ATS

16 specialist prompts across momentum, pace, shooting, injuries, coaching, and market perception. Bonus first-round game data for ChatGPT and Claude included.

Launch price ($9) ends before NCAA First Round tips on March 19, 2026 —

How It Works

Pick a matchup or use the included first-round slate

Start with a ready game, then fill only the missing lines.

Paste one agent into ChatGPT, Claude, or Gemini

The prompt structures the analysis and returns a clear read for that angle.

Run more angles when you want more confidence

Use one agent for a quick check, or all 16 for the full matchup read.

Optional Pro Mode (included): batch-run all 16 agents

Most buyers use Standard Mode. If you're technical, Pro Mode lets you run one matchup file through all 16 agents and export a summary in one pass.

Show technical setup (YAML + one command)
  • One matchup file for consistent inputs across all agents
  • One command to generate all 16 outputs
  • One summary table for fast pregame review

1) Copy the example config

cp configs/matchup.example.yaml configs/matchup.my-game.yaml

2) Add your model key locally

export AGENT_RUN_PROVIDER=anthropic
export ANTHROPIC_API_KEY=your_key_here
export ANTHROPIC_MODEL=claude-sonnet-4-6

3) Run the full batch

npm run run:agents -- --config configs/matchup.my-game.yaml

Not technical? Skip this section. Standard Mode is the default and is fully supported.

FAQ

What LLMs do these work with?

ChatGPT, Claude, Gemini, and other LLMs. They're plain-text prompts, not software. If it accepts a prompt, it works.

Do I need a paid AI subscription?

Free tiers work fine. Paid tiers like GPT-4 or Claude Pro give sharper results on the more complex agents, but they're not required.

Where do I get the game data?

Free sources — Barttorvik, ESPN, Sports Reference. A data-sourcing guide and a copy-paste template are included in the zip.

What exactly do I get?

16 precision-engineered prompt files (one per angle), a data-sourcing guide with free stat sources, a copy-paste input template, and a README. Instant email delivery. No software to install, no account to create.

Is this a picks service?

No. This is a structured research workflow. It helps you evaluate matchups faster and more consistently, but final decisions are yours.

What if it doesn't work for me?

If you run an agent and the output isn't structured and useful, email us within 7 days and we'll refund it.

What is Pro Mode?

Pro Mode is an included automation layer for technical users. You can run all 16 agents from one YAML matchup file and get a single summary report. It is not required, and Standard Mode remains the primary experience.

First-round matchups are loaded now. Open the runner before the board moves.

Run first-round matchups now