.skills/test-loop/SKILL.md
Continuously run tests, create failure cards, and delegate fixes to /implement until the suite is fully green. Uses ralph to prevent stopping between iterations.
npx skillsauth add swissarmyhammer/swissarmyhammer test-loopInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Continuously run tests and fix failures until the entire suite is green.
This skill is an orchestrator. It does not write code or run tests itself. It delegates to /test (which finds failures and creates kanban cards) and /implement (which picks up those cards and fixes them), and uses ralph to stay alive between iterations.
ralph with op: "set ralph" and instruction "Run tests and fix failures until all green"./test — it runs the suite, creates test-failure kanban cards for any failures.kanban for tasks tagged test-failure./implement once — it picks up exactly one card and fixes it. Then go back to step 2 to re-run tests and check for remaining failures. Repeat until no failure cards remain.test-failure — if any exist, continue the loop regardless of what /test reported. Only when the kanban board has zero test-failure cards and /test reports a fully green run (zero failures, zero warnings, zero skipped) may you clear ralph and report.ralph with op: "clear ralph" when both conditions are met:
test-failure (verified by a direct kanban query)./test reports all green with zero failures, zero warnings, and zero skipped tests./test's prose summary — always re-check the kanban board independently before clearing ralph./test owns test execution, failure analysis, and card creation. Do not run tests yourself./implement owns card pickup and code fixes. Do not write code yourself./implement reports it is stuck on a card, skip it and continue the loop — do not try to fix it yourself.research
Create a single, well-researched kanban task. Use when the user wants to add a task, track an idea, or capture work without entering full plan mode.
testing
Drive kanban tasks from ready to done by looping implement → test → review until each task is clean. Supports single-task mode (one task id) and scoped-batch mode (all ready tasks in a tag/project/filter). Uses ralph to prevent stopping between iterations.
tools
Use when starting any conversation - establishes how to find and use skills, requiring Skill tool invocation before ANY response including clarifying questions
testing
Run tests and analyze results. Use when the user wants to run the test suite or test specific functionality. Test runs produce verbose output — automatically delegates to a tester subagent.