2
GitHub Stars
1
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill 89jobrien/steve --skill tdd-pytest- SKILL.md4.7 KB
Overview
This skill is a Python/pytest TDD specialist that guides test-driven development workflows and automates pytest tasks. It helps write failing-first tests, implement minimal code to pass them, and refactor while keeping test suites green. It also audits test quality, runs tests with coverage via uv, and generates a TESTING_REPORT.local.md report.
How this skill works
The skill inspects project layout, tests, and pyproject.toml to suggest or create tests and pytest configuration. It can produce failing tests (RED), implement minimal fixes (GREEN), and propose refactors while re-running pytest after each change. It runs tests through uv, collects coverage, and compiles an audit and evidence into TESTING_REPORT.local.md.
When to use it
- Starting a new feature using red-green-refactor TDD
- Improving or auditing an existing pytest suite for quality issues
- Configuring pytest and coverage in pyproject.toml
- Running tests and coverage via uv and generating a test report
- Automating test runs and regression checks before merges
Best practices
- Write a focused failing test first; ensure failure is for the intended reason
- Keep tests small and descriptive; name files test_*.py or *_test.py and functions test_*
- Use fixtures for shared setup and descriptive names (e.g., mock_client, sample_user)
- Parametrize repetitive scenarios and use async markers for async functions
- Run pytest frequently during refactor steps and prefer minimal implementations to pass tests
Example use cases
- Create a failing unit test for a new function and implement the smallest passing code
- Audit tests to find flaky cases, missing assertions, and coverage gaps then produce remediation steps
- Add asyncio support and configure test discovery in pyproject.toml
- Run uv run pytest --cov and generate TESTING_REPORT.local.md summarizing results and recommendations
- Add parametrized tests for input/output combinations and convert duplicated setups into fixtures
FAQ
I follow TDD: I write the failing test first, then suggest minimal production code to make it pass, and finally recommend refactors. You can accept or modify the implementation steps.
How is the TESTING_REPORT.local.md structured?
It includes a test execution summary, per-module coverage metrics, audit findings with severity and file:line references, recommendations, and command output evidence.