quell check
Reads docstrings, Pydantic models, and type annotations from a file or directory, extracts every testable requirement, checks which ones have no test, and optionally generates a verified test for each gap.
Usage
quell check [TARGET] [OPTIONS]
Arguments
| Argument | Default | Description |
|---|---|---|
TARGET | . | File or directory to scan |
Options
| Option | Default | Description |
|---|---|---|
--fix | False | Generate and write verified tests for all gaps |
--sources | docstring,type | Comma-separated spec sources to use |
--root | . | Project root directory |
Examples
# Scan a file and show gaps
quell check src/payments.py
# Scan a directory
quell check src/
# Generate and write verified tests for all gaps
quell check src/payments.py --fix
# Use only docstring specs
quell check src/ --sources docstring
# Use both docstring and type specs (default)
quell check src/ --sources docstring,type
Console output
Requirements — src/payments.py
Function Kind Description Covered
process_payment MUST_RAISE ValueError: If amount is zero... NO
process_payment MUST_RETURN dict with transaction_id, statu... YES
process_payment BOUNDARY Amount must be greater than 0 NO
PaymentRequest ENUM_VALID currency: USD, EUR, GBP NO
PaymentRequest BOUNDARY amount: gt=0 YES
Score: 40% (2/5 covered)
3 gap(s) found. Run with --fix to generate tests.
The --fix flag
When --fix is passed, Quell generates a candidate test for each uncovered requirement and verifies it in two phases:
- The test must PASS on the original (correct) code.
- The test must FAIL on code that violates the requirement.
Only tests that pass both phases are written to disk.
Spec sources
| Source | What it reads |
|---|---|
docstring | Google-style Raises:, Returns:, and Args: blocks; boundary phrases |
type | Pydantic Field validators; Literal type annotations |
Use --sources to restrict which readers run. By default both docstring and type are active.