Testing strategy

Skill frobinson47/solo-dev-suite/skills/testing-strategy

Generates a right-sized, stack-aware testing strategy specifying what to unit test, integration test, E2E test, and manually test -- avoiding both under-testing and 100%-coverage dogma.From its SKILL.md

Install

npx -y skills add frobinson47/solo-dev-suite --skill testing-strategy

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

4 stars4 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

2.6 KB, as published. Nobody here has run it

testing-strategy

When to use

Starting a new project and need a testing plan.
Reviewing whether your current testing approach still makes sense.
Before a launch push to ensure critical paths have coverage.

When NOT to use

For generating actual test code (this produces the strategy doc, not tests).
For running tests or checking coverage.

Prerequisites

Project must be onboarded in solo-dev-suite.

Methodology

Coverage targets are qualitative, not percentages

"High" means the high-risk modules are well covered. Not a number. Goodhart's law applies.

Four categories

Unit (30-50% effort): Pure business logic, math, validation, parsers. Skip UI glue.
Integration (30-40% effort): The high-leverage tier. DB <-> API, auth flows, payment flows.
E2E (10-20% effort): Only critical conversion paths. One flaky E2E test is worse than none.
Manual (10-20% effort): Everything else. Documented so "manual" doesn't mean "skipped."

Explicit skips matter

A strategy that says "don't bother testing UI components" is useful. Silence on UI components is ambiguous.

Staleness detection

The review_cadence field (monthly/quarterly/biannual/annual) drives staleness warnings on show.

Operations

design

Create the initial testing strategy.

echo '{ full payload }' | python scripts/testing_tool.py design <slug> --from-stdin

show

Display strategy with staleness check.

python scripts/testing_tool.py show <slug>
python scripts/testing_tool.py show <slug> --category unit
python scripts/testing_tool.py show <slug> --json

iterate

Update the strategy (requires reason).

echo '{"reason":"Adding Stripe, need integration tests","categories":{"integration":{...}}}' \
  | python scripts/testing_tool.py iterate <slug> --from-stdin

review

Mark strategy as reviewed, resets staleness timer.

python scripts/testing_tool.py review <slug>

render

Generate TESTING_STRATEGY.md.

python scripts/testing_tool.py render <slug>

delete

Remove sidecar.

python scripts/testing_tool.py delete <slug> --yes

Files

File	Purpose
`scripts/testing_tool.py`	CLI with all subcommands
`templates/testing.schema.json`	Sidecar JSON Schema
`templates/TESTING_STRATEGY.md.tmpl`	Rendered doc template