How It Works

How It Works

Index Inspector bridges the gap between your source code and your database performance. Here is the high-level flow of how it processes your project:

1. User Sends Command: You interact with the bot via Telegram (e.g., /scan main).
2. Bot Receives Request: The bot validates your project settings and repository access.
3. Code Analysis: Our backend pulls the specified branch, parses the codebase, and identifies all database interaction points (SQL queries, ORM methods).
4. Index Optimization Logic: The analyzer evaluates these queries against best practices and common indexing patterns.
5. Recommendation Generation: The bot compiles a list of recommended indexes per table and column.
6. Response Returned: You receive a summary of the scan and a detailed result set directly in Telegram.

Under the Hood

Index Inspector is built for reliability and speed. When you run a scan, we use optimized static analysis tools that look for patterns in your code. Instead of running your code, which could be slow or risky, we analyze the structure of your queries.

• Privacy & Data Handling: We only scan the paths you specify. We do not store your source code; it is processed in memory and discarded after the analysis is complete.
• Accuracy: By scanning your migration files (via the specified Migration Scan Paths), we extract your current database schema and identify all columns that are already indexed. This ensures that our recommendations only include new optimizations that aren’t already implemented in your database.
• Reliability: Our engine is designed to handle various frameworks (like Laravel, Django, Express) and databases (PostgreSQL, MySQL), ensuring that the recommendations are tailored to your specific tech stack.