AgentSkillsCN

pytest-workflow

Pytest测试设计工作流程。当测试或证据产物作为验证的主要输出时使用；切勿用于产品需求的优先级排序或架构拓扑的选择。

SKILL.md
--- frontmatter
name: pytest-workflow
description: "Pytest test design workflow. Use when test or evidence artifacts are the primary output for verification; do not use for product requirement prioritization or architecture topology selection."

Pytest Workflow

Trigger Boundary

  • •Use when Python verification is implemented with pytest.
  • •Do not use for framework-agnostic test strategy only; use testing-* skills.
  • •Do not use for load/performance benchmarking scripts.

Goal

Create fast, reliable, and debuggable pytest suites aligned with change risk.

Inputs

  • •Change scope and risk profile
  • •Existing fixture and test structure
  • •Environment and data setup constraints

Outputs

  • •Test case matrix with fixture strategy
  • •Parametrized tests for boundary and error scenarios
  • •Failure triage notes with reproducible commands

Workflow

  1. •Map behavior risks to unit/integration pytest scopes.
  2. •Build reusable fixtures with explicit lifecycle.
  3. •Use parametrization for edge-case coverage without duplication.
  4. •Keep assertions specific and failure messages actionable.
  5. •Optimize runtime by isolating slow tests and markers.

Quality Gates

  • •Fixtures are deterministic and avoid hidden global state.
  • •Parametrization improves coverage without unreadable tests.
  • •Test failures point to a single clear cause.
  • •Test runtime remains practical for local and CI execution.

Failure Handling

  • •Stop when fixtures produce non-deterministic outcomes.
  • •Escalate when test runtime blocks practical feedback loops.