Flask patterns -- application factory, blueprints, error handlers, extensions, request lifecycle, configuration, logging, CLI commands
98
98%
Does it follow best practices?
Impact
98%
1.28xAverage score across 5 eval scenarios
Passed
No known issues
{
"context": "Tests whether the agent follows Flask best practices when building a bookstore inventory API with SQLite, rate limiting, and proper error handling.",
"type": "weighted_checklist",
"checklist": [
{
"name": "create_app factory",
"description": "app/__init__.py defines a create_app() function that creates and returns a Flask app -- not a module-level app",
"max_score": 12
},
{
"name": "Blueprints in separate files",
"description": "Books and categories routes are in separate Blueprint files under app/routes/ -- not all routes in one file or on the app directly",
"max_score": 10
},
{
"name": "Extensions init_app pattern",
"description": "CORS and rate limiter are instantiated in extensions.py without an app argument and bound via init_app(app) inside create_app()",
"max_score": 10
},
{
"name": "Custom exception hierarchy",
"description": "Agent defines AppError base class with NotFoundError, ValidationError subclasses -- routes raise these rather than returning error dicts manually",
"max_score": 10
},
{
"name": "Consistent error envelope",
"description": "All error responses (validation, not-found, server error) return {\"error\": {\"code\": \"...\", \"message\": \"...\"}} -- no stack traces or inconsistent shapes",
"max_score": 10
},
{
"name": "Generic exception handler logs errors",
"description": "The catch-all Exception handler calls app.logger.exception() or similar logging before returning a 500 with a generic message",
"max_score": 8
},
{
"name": "DB teardown with teardown_appcontext",
"description": "Database connection cleanup uses teardown_appcontext (not after_request) to ensure connections are closed even on errors",
"max_score": 8
},
{
"name": "DB connection via flask.g",
"description": "Database connections are stored on flask.g (g.db) -- not in module-level globals",
"max_score": 8
},
{
"name": "Config supports testing",
"description": "create_app() accepts a config parameter and the testing config uses an in-memory SQLite database (':memory:')",
"max_score": 8
},
{
"name": "Error handlers registered in factory",
"description": "Error handlers are registered centrally via register_error_handlers(app) inside create_app() -- not per blueprint",
"max_score": 8
},
{
"name": "Test fixtures",
"description": "tests/conftest.py creates the app via create_app() with testing config and provides app and client fixtures",
"max_score": 8
}
]
}