Smart Test reference

Find coverage gaps in your codebase and generate pytest tests with edge cases, boundary values, and error paths.

Classes

Class Description
ModuleCoverage Coverage data for a single module.
TestAuditWorkflow Autonomous test coverage audit with Agent SDK subagents.
ASTFunctionAnalyzer AST-based function analyzer for accurate test generation.
FunctionSignature Detailed function analysis for test generation.
ClassSignature Detailed class analysis for test generation.
TestGenerationWorkflow SDK-native test generation with three specialized subagents.
TestGenerationTask Tracks the state of a single test generation task.
ParallelTestGenerationWorkflow Generate and complete behavioral tests in parallel using multi-tier LLMs.

ModuleCoverage fields

ModuleCoverage is a dataclass.

Field Type Default
path str —
stmts int —
covered int —
missing_lines list[int] field(default_factory=list)
pct float 0.0
priority float 0.0

FunctionSignature fields

FunctionSignature is a dataclass.

Field Type Default
name str —
params list[tuple[str, str, str | None]] —
return_type str | None —
is_async bool —
raises set[str] —
has_side_effects bool —
docstring str | None —
complexity int 1
decorators list[str] field(default_factory=list)

ClassSignature fields

ClassSignature is a dataclass.

Field Type Default
name str —
methods list[FunctionSignature] —
init_params list[tuple[str, str, str | None]] —
base_classes list[str] —
docstring str | None —
is_enum bool False
is_dataclass bool False
required_init_params int 0

TestGenerationTask fields

TestGenerationTask is a dataclass.

Field Type Default
module_path str —
coverage float —
output_path str —
status str 'pending'

TestAuditWorkflow methods

Method Parameters Returns Description
__init__ self, *, system_prompt_suffix: str = '', **kwargs: Any None Initializes the workflow with an optional system prompt suffix.
execute self, **kwargs: Any WorkflowResult Runs the coverage audit and returns a structured result.

ASTFunctionAnalyzer methods

Method Parameters Returns Description
__init__ self — Initializes the analyzer.
analyze self, code: str, file_path: str = '' tuple[list[FunctionSignature], list[ClassSignature]] Parses source code and returns extracted function and class signatures.
visit_FunctionDef self, node: ast.FunctionDef None Visits a function definition node during AST traversal.
visit_AsyncFunctionDef self, node: ast.AsyncFunctionDef None Visits an async function definition node during AST traversal.
visit_ClassDef self, node: ast.ClassDef None Visits a class definition node during AST traversal.

TestGenerationWorkflow methods

Method Parameters Returns Description
__init__ self, **kwargs: Any — Initializes the workflow.
execute self, **kwargs: Any WorkflowResult Runs test generation with three specialized subagents and returns a structured result.

ParallelTestGenerationWorkflow methods

Method Parameters Returns Description
default_context cls, xml_config: dict | None = None WorkflowContext Returns the default workflow context, optionally configured with XML.
discover_low_coverage_modules self, top_n: int = 200 list[tuple[str, float]] Returns the top N modules ranked by lowest coverage.
analyze_module_structure self, file_path: str dict[str, Any] Analyzes a module's structure and returns a summary dict.
generate_test_template_with_ai self, module_path: str, structure: dict[str, Any] str Generates a test file template for the given module using an LLM.
complete_test_with_ai self, template: str, module_path: str str Completes a test template into a full test file using an LLM.
process_module_batch self, modules: list[tuple[str, float]], output_dir: Path, batch_size: int = 10 list[TestGenerationTask] Processes a batch of modules and returns generation task results.
execute self, top: int = 200, batch_size: int = 10, output_dir: str = 'tests/behavioral/generated', **kwargs WorkflowResult Runs parallel test generation across low-coverage modules and returns a structured result.

Functions

Function Parameters Returns Description
parse_coverage_json json_path: str list[ModuleCoverage] Parse pytest-cov's coverage.json output.
prioritize_modules modules: list[ModuleCoverage], min_threshold: float = 50.0 list[ModuleCoverage] Sort modules by priority and filter below threshold.
group_into_batches modules: list[ModuleCoverage], max_batches: int = 5 list[dict] Group modules into batches by subsystem (package path).
format_test_gen_report result: dict, input_data: dict str Format test generation output as a human-readable report.
generate_test_for_function module: str, func: dict str Generate executable tests for a function based on AST analysis.
generate_test_cases_for_params params: list dict Generate test cases based on parameter types.
get_type_assertion return_type: str str | None Generate assertion for return type checking.
get_param_test_values type_hint: str list[str] Get test values for a single parameter based on its type.
generate_test_for_class module: str, cls: dict str Generate executable test class based on AST analysis.
main — None CLI entry point for test generation workflow.

Raises

Function Exception Message
parse_coverage_json FileNotFoundError 'Coverage file not found: {...}'
parse_coverage_json ValueError 'Invalid coverage JSON at {...}: {...}'
parse_coverage_json ValueError "Unexpected coverage JSON structure in {...}: 'files' key missing or not a dict"

get_param_test_values return values

get_param_test_values returns a list[str]. The list may contain the following literal value:

Value
"test_value"

Constants

Constant Type Members
AUDIT_SYSTEM_PROMPT str System prompt for the coverage-auditor subagent. Instructs it to review module prioritization and return a JSON object with a "modules" key.
PLAN_SYSTEM_PROMPT str System prompt for the test-planner subagent. Instructs it to group modules into logical batches and return a JSON object with a "batches" key, each containing a "task_xml" field.
BATCH_TASK_TEMPLATE str XML task prompt template. Interpolation keys: {batch_id}, {subsystem}, {target_pct}, {missing_lines_summary}, {source_path}, {key_signatures}, {test_path}, {test_class_specs}, {module}.
DEFAULT_SKIP_PATTERNS list '.git', '.hg', '.svn', 'node_modules', 'bower_components', 'vendor', '__pycache__', '.mypy_cache', '.pytest_cache', '.ruff_cache', '.hypothesis', 'venv', '.venv', 'env', '.env', 'virtualenv', '.virtualenv', '.tox', '.nox', 'build', 'dist', 'eggs', '.eggs', 'site-packages', '.idea', '.vscode', 'migrations', 'alembic', '_build', 'docs/_build'
_SUBAGENT_NAMES (audit) list 'coverage-auditor', 'gap-analyzer', 'test-planner'
_SUBAGENT_NAMES (gen) list 'function-identifier', 'test-designer', 'test-writer'

Source files

Tags

tests, coverage, generation