Executes testing and regression workflows with a focus on boundary cases and evidence-based verification.

Install

mkdir -p .claude/skills/test-jacksmack1971 && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/14552" && unzip -o skill.zip -d .claude/skills/test-jacksmack1971 && rm skill.zip

Installs to .claude/skills/test-jacksmack1971

Activation

This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.

Run DegenExus pytest planning and regression workflow with Prove-It, FSV-AAA, and edge-case evidence.
101 charsno explicit “when” trigger
Advanced

Key capabilities

  • Identify the source of truth and current test patterns
  • Write or describe failing regression tests for bugs
  • Apply FSV-AAA for test structure
  • Cover boundary and equivalence classes in tests
  • Mock LLM providers, network calls, time, and randomness
  • Run targeted pytest then broader regression tests

How it works

The skill identifies test patterns, writes failing regressions for bugs, and applies FSV-AAA for test structure. It covers boundary cases and mocks external dependencies before running tests.

Inputs & outputs

You give it
Test scope or behavior
You get back
Pytest planning and regression workflow execution with evidence fields

When to use test

  • Writing regression tests
  • Verifying bug fixes
  • Auditing edge cases

About this skill

Test Workflow

test-engineer is the canonical owner for DegenExus test authoring and coverage analysis. test-writer is a read-only deprecated alias and must not be routed for new work.

Required References

  • .claude/rules/synergy-contract.yml
  • .claude/rules/evidence-schema.yml
  • .claude/skills/test-regression/SKILL.md
  • .claude/skills/fsv-verify/SKILL.md
  • .claude/skills/edge-case-audit/SKILL.md

Steps

  1. Identify the source of truth and current test patterns.
  2. For bugs, write or describe the failing regression first and record failing_before.
  3. Use FSV-AAA where practical: PRE source-of-truth read, one ACT, POST reread, expected DIFF assertion.
  4. Cover boundary/equivalence classes: empty/zero, min/max boundary, malformed/adversarial input, and ordering/concurrency where applicable.
  5. Use pytest-mock's mocker fixture; do not import unittest.mock directly.
  6. Mock LLM providers, yfinance/network calls, time, and randomness.
  7. Run targeted pytest first, then broader regression tests when feasible.
  8. Emit .claude/rules/evidence-schema.yml fields with edge_cases, failing_before, and passing_after evidence when applicable.

When not to use it

  • When `test-writer` is routed for new work

Limitations

  • The skill does not use `test-writer` for new work.
  • The skill requires documentation of edge cases and failing/passing states.
  • The skill uses `pytest-mock`'s `mocker` fixture, not `unittest.mock` directly.

How it compares

This workflow systematically structures tests with FSV-AAA and covers specific boundary conditions, unlike ad-hoc testing.

Compared to similar skills

test side by side with the closest alternatives in the catalog.

SkillInstallsUpdatedSafetyDifficulty
test (this skill)02moNo flagsAdvanced
python-testing-patterns772moReviewIntermediate
python-playground24moReviewBeginner
mflux-manual-testing22moReviewBeginner

Try saying

Example prompts that trigger this skill in your AI assistant.

Search skills

Search the agent skills registry