Use local ahent with a Supabase MCP client to solve 4 progressive data visualization challenges.
workshop-04/
├── exercise_0/ # Introductory exercise
├── exercise_1/ # Era Batting Average Trends (Line Plot)
├── exercise_2/ # Top Home Run Hitters Heatmap
├── exercise_3/ # Home Run Distribution Box Plot
├── exercise_4/ # Team Performance Correlation Matrix
├── src/mlb/ # Database utilities & connection tools
├── outputs/ # Save your visualizations here
└── pyproject.toml # Python dependencies (uv managed)
- Environment Setup: Ensure you have
uvpackage manager installed.
pip install uv
-
Database Connection: Configure your Supabase connection via MCP. For that, add the MCP definition in
.cusros/mcp.json. The access token will be provided in the meeting chat. -
Rules definition: Complete the plotting guidelines rool in
.cursor/rules/plotting-guidelined.mdc. Create any extra rool you consider necessary. -
Commands definition: Complete the database MCP command in
.cursor/commands/database.md. Create any extra command you consider necessary.
| Resource | Purpose | Link |
|---|---|---|
| Supabase MCP | Database connection & queries | Setup Guide |
| Cursor Commands | AI assistant capabilities | Command Reference |
- Start simple: Create a simple version of the rules and commands. Try working on exercise 1 and iterate as you identify problems with your rules.
- Use meta promting: Use AI to generate the rules, commands and prompts, but only once you know that do you want to include on each one.
- Breakdown problems: Separate each exercise into smaller pieces that you can iterate. Don't try to one-shot every problem. Use change of thought (CoT) in your promting.