Instruction file imported from vitolytics/tennis-data-pipeline (.github/instructions/AGENT.instructions.md). Copyright stays with the author.

Agent Instructions for InfoTennis Project

Database Table Naming Convention

When creating or referencing database tables in this project, follow this naming structure:

Format: {source}_{category/grain}_{subject}

Components:

1. source: The actual source where data is scraped from (e.g., atpcom, wtacom), not the underlying data provider
2. category/grain: The granularity or category level (e.g., match, player, tournament, point)
3. subject: The specific data subject or type (e.g., matchinfo, rallyanalysis, keystats, courtvision, strokes)

Examples:

• atpcom_matchinfo - Match information from atp.com
• atpcom_match_rallyanalysis - Match-level rally analysis from atp.com
• atpcom_match_keystats - Match-level key statistics from atp.com
• atpcom_match_courtvision - Match-level court vision data from atp.com
• atpcom_match_strokes - Match-level stroke analysis from atp.com
• atpcom_tournament_calendar - Tournament calendar information from atp.com
• atpcom_point_data - Point-level data from atp.com (even if powered by Infosys)

Guidelines:

• Use lowercase with underscores (snake_case)
• Be descriptive but concise
• Maintain consistency with the three-part structure
• When the grain is implicit or redundant, it may be omitted (e.g., atpcom_matchinfo rather than atpcom_match_matchinfo)