Testing Skill
Use this skill as the cross-stack validation orchestrator.
Scope
- •Define validation gates and execution order
- •Select stack commands from canonical stack skills
- •Standardize failure triage and rerun strategy
- •Produce structured validation output for handoff/gates
Canonical Stack Command Sources
- •Python:
.agent/skills/python/SKILL.md - •Node:
.agent/skills/node/SKILL.md - •Go:
.agent/skills/go/SKILL.md - •Rust:
.agent/skills/rust/SKILL.md
For deep Python testing setup and pytest ecosystem details, load:
- •
.agent/skills/python/pytest/SKILL.md
Validation Gate Workflow
- •Detect stack from lock/config files.
- •Run format checks.
- •Run lint/static analysis.
- •Run type/compile checks when applicable.
- •Run tests and coverage when required.
- •If any check fails, triage root cause and rerun required scope.
- •Finish only when gate criteria pass.
Failure Triage Policy
- •Reproduce with smallest failing scope first.
- •Use deterministic reruns (
--lf, targeted test path, targeted package/module). - •Capture concise failure tails and stack trace context.
- •Avoid broad retries before identifying root cause.
Output Contract
json
{
"validation_passed": true,
"stack": "python|node|go|rust",
"format_status": "pass|fail|skipped",
"lint_status": "pass|fail|skipped",
"typecheck_status": "pass|fail|skipped",
"test_status": "pass|fail|skipped",
"coverage": {
"lines": 0,
"branches": 0
},
"commands_run": [
"<format>",
"<lint>",
"<typecheck>",
"<test>"
],
"notes": "root cause, reruns, and residual risk"
}
Guardrails
- •Never mark gate as passed with failing mandatory checks.
- •Keep local validation aligned with CI.
- •Keep command source-of-truth in stack skills.