Test suites + deterministic Python for your agent skills' decision logic. Adversarial persona reviewers write the tests; the generated code must keep passing them — zero LLM calls at inference. Try: uvx temper-skills audit <skill.md>
Give an agent skill's decision logic a test suite and a deterministic implementation — a labeled validation dataset plus versionable Python — via an adversarial multi-persona loop that runs natively in Claude Code: proposer plus persona subagents through the Task tool, entirely on your Claude Code subscription, no…
Frozen, deterministic decision (no LLM): maps painmalleolarzone, bonetendernesslateralmalleolus, bonetendernessmedialmalleolus, canbearweight, visibledeformity, spraingrade, hourssinceinjury, ageyears, patientprofile to one of urgentcare, seekimaging, seeclinician, policeacute, policesubacute. Use when this decision…
Tempered orchestrator: chains frozen, deterministic decision trees (decidewalk, decidemeal, decidevet) and keeps only the generative note. Use to run a daily dog-care flow with each decision made by code (no LLM) and only the prose left to the model.
Frozen, deterministic decision (no LLM): maps fooditem, foodform, dogweightkg, dogbreed, quantitygrams to one of no — toxic to dogs, never feed, no — concentrated form multiplies toxin and fat dose per gram, no — low-fat/sugar-free peanut butter may contain xylitol; check the label, caution — large dose relative to…
Frozen, deterministic decision (no LLM): maps projectlicense, dependencylicense, linking, distributing, modifieddependency to one of okinternaluse, mustoffersource, projectmustbegpl, compatible, incompatible, reviewrequired. Use when this decision must be made consistently and auditably — extract the features, call…
Frozen, deterministic decision (no LLM): maps zonetype, day, hour, ispublicholiday, hasresidentpermit to one of prohibited, noparkingstreetcleaning, payanddisplay2h, free, permitrequired. Use when this decision must be made consistently and auditably — extract the features, call canipark(), and relay its verdict…
Frozen, deterministic decision (no LLM): maps priority, securityscore, customertier, category, slabreached to one of humanreview, escalatesecurity, escalateurgent, routebilling, routeengineering, routedefault. Use when this decision must be made consistently and auditably — extract the features, call routeticket()…
★not rated 4 26d agoA80 tokens
originalApache-2.0
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: