Ling0925
/
buster
mirror of https://github.com/buster-so/buster.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

More Prompt fixes for math logic errors

This commit is contained in:
Jacob Anderson 2025-07-10 15:18:48 -06:00
parent 4d55367d47
commit 8a22c2eefe
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
packages/ai/src/agents/think-and-prep-agent/think-and-prep-instructions.ts
View File

@ -234,7 +234,7 @@ Once all TODO list items are addressed and submitted for review, the system will
- **CRITICAL: Apply <join_and_calculation_best_practices> when planning SQL queries involving joins or calculations** - Always analyze how joins will affect your data and calculations, especially for percentage calculations, many-to-one relationships, and rate calculations. Review your SQL logic to prevent common issues like denominator bias, data duplication, and incorrect aggregation levels.
- After evaluating precomputed metrics, ensure your approach still adheres to <filtering_best_practices>, <aggregation_best_practices>, and <join_and_calculation_best_practices>.
- When building bar charts, Adhere to the <bar_chart_best_practices> when building bar charts. **CRITICAL**: Always configure axes as X-axis: categories, Y-axis: values for BOTH vertical and horizontal charts. Never swap axes for horizontal charts in your thinking - the chart builder handles the visual transformation automatically. Explain how you adhere to each guideline from the best practices in your thoughts.
- Whenever you are using a percentage column, use the \`executeSql\` tool to determine if the percentage column is stored as a percentage or a decimal (e.g. 50 vs .5). If you need to do a calculation, make sure to use the \`executeSql\` tool to understand how the numbers are stored and then use the correct aggregation function.
- Everytime you are using a percentage column, use the \`executeSql\` tool to determine if the percentage column is stored as a percentage or a decimal (e.g. 50 vs .5). If you need to do a calculation, make sure to use the \`executeSql\` tool to understand how the numbers are stored and then use the correct aggregation function.
- **CRITICAL**: Adhere to the <sql_best_practices> when building SQL queries to ensure there are no errors in SQL building that would cause the user to be mislead.
- **MANDATORY SQL VALIDATION PROTOCOL**: For every SQL query you plan, you MUST:
1. **Pattern Check**: State which high-risk patterns (if any) your query matches from <join_and_calculation_best_practices>