--- name: slot-machine description: Use when a well-specified…
Community ralph loop for testing, sourced from github. Verified exit condition, evaluator-gated.
--- name: slot-machine description: Use when a well-specified task has meaningful design choices and you want to maximize quality by comparing multiple independent attempts. Works for coding, writing, and custom task types. Triggers on "slot-machine", "best-of-N", "pull the lever", "parallel implementations", or when quality matters more than speed and the spec is clear enough for independent work. --- # Slot Machine Best-of-N parallel implementation for any task type. Run N independent attempts at the same spec in parallel. Review each. Pick the best — or synthesize the best elements into a single winner. Core principle: LLMs are probabilistic. More attempts = better outcomes. Trade compute for quality. Announce at start: "I'm using the slot-machine skill ({profile name} profile) to run N parallel implementations." ## What This Is NOT Standard multi-agent patterns split DIFFERENT tasks across agents (frontend, backend, tests in parallel). Every major tool does this — it's table stakes. Slot-machine gives the SAME spec to N agents and compares their FULL attempts. The value isn't parallelism — it's competition and selection. Each slot is an independent attem
claude-code
More testing loops
Refactor in tested checkpoints
Reshape the architecture in small, independently reviewed steps — live-test, commit a checkpoint, and log progress each pass.
Refactor until you are happy with the architecture. After each significant step, live-test the system, run autoreview, and commit. Track progress in /tmp/refactor-{projectname}.md.
testingmedium risk
Kill flaky tests
Run your test suite repeatedly, collect every intermittent failure, and fix or quarantine flaky tests until you get five consecutive green runs.
/loop run my test suite 20 times, collect every intermittent failure, fix or quarantine the flaky ones, and don't stop until you get 5 consecutive fully-green runs
testinghigh risk
Cursor "Iterate Until Tests Pass, Never Touch the Tests"
First-party Cursor guidance for the iterate-until-green loop, with the key anti-reward-hacking clause: the agent may never modify the tests it is trying to satisfy. Works in Cursor, Claude Code /goal, and Codex.
"Write code that makes these tests pass. Do NOT modify the tests. Keep iterating — run the suite, fix failures, run again — until all tests pass." (paraphrase of Cursor's official agent best-practices guidance)
testingmedium risk