agtest 0.2.0

AI-powered tool to automatically generate compliant unit tests for FastAPI controllers and service layers, enforcing strict project standards.

Purpose: The Agent AI tool automatically analyzes FastAPI controller/service files and generates correct unit tests based on strict project rules.

Core Features

-Detect whether the input file is a controller or service -Generate FastAPI test cases with mandatory project standards: -Mock DB Session (MagicMock(spec=Session)) -TestClient with middleware injecting request state -CustomException handler setup -Dependency override for get_db -Cleanup overrides after each test -Ensure success + negative case per endpoint -Follow session-per-request mocking pattern -Handle async + TaskGroup errors (ExceptionGroup) gracefully

Flow- agent_ai -> detect_file_type -> detect_controller_prefix -> run_agent detect_controller_prefix -> get_prefix_for_controller run_agent -> build_prompt -> get_model_client -> generate_response

build_prompt -> extract_schemas_and_modules -> extract_schema_metadata_from_ast

extract_schema_metadata_from_ast -> find_schema_file -> parse_ast_file -> get_schema_type -> extract_fields -> extract_validators

Configuration- It should have MODEL_NAME , MAX_TOKENS , TEMPERATURE , GOOGLE_API_KEY. To change output dir go in run_agent and change it.

All files will be kept at same level in a folder.

Functions list agent-ai.py :- agent_ai agent-services.py :- get_model_client , generate_response , build_prompt, run_agent utilities.py :- extract_schema_metadata_from_ast , find_schema_file , parse_ast_file , get_schema_type ,extract_fields , extract_validators , extract_schemas_and_modules , detect_controller_prefix , get_prefix_for_controller , detect_file_type