How prooflinedev/ai-review-benchmarks works
This repository is a small report-scheduling and delivery service, provided with a CLI and an AWS Lambda adapter. A user schedules reports, processes due deliveries, and inspects history, exports, or summary counts. The shared reports module owns those operations; SQLite holds delivery and receipt state, with numbered SQL migrations. The README also describes the benchmark and deployment exercise, but the graph focuses on the implemented runtime. Relationships are limited to calls and database operations shown in the excerpts; the migration’s schema details were not sampled.
Main entry points: reports.py Report Service · console.py CLI Console · handler.py Lambda Handler · 001_initial.sql Initial Migration
Components 11
Service Interfaces
- CLI Console
console.py - Lambda Handler
aws/handler.py
Report Operations
- Report Service
reports.py - Scheduling
reports.py - Due Delivery
reports.py - History Paging
reports.py - CSV and Archive
reports.py - Delivery Summary
reports.py
Persistence
- SQLite Database
- Initial Migration
migrations/001_initial.sql
Other components
- Operator
Connections 12
- Operator → CLI Console: runs commands
console.py - CLI Console → Report Service: calls operations
console.py - Lambda Handler → Report Service: calls operations
aws/handler.py - Report Service → Scheduling: schedules jobs
reports.py - Report Service → Due Delivery: processes due jobs
reports.py - Report Service → History Paging: pages history
reports.py - Report Service → CSV and Archive: writes or batches
reports.py - Report Service → Delivery Summary: counts deliveries
reports.py - Report Service → SQLite Database: reads and writes
reports.py - Report Service → Initial Migration: loads SQL migrations
reports.py - CLI Console → SQLite Database: opens SQLite
console.py - Lambda Handler → SQLite Database: opens in-memory SQLite
aws/handler.py