#SkillRepo stars
- 12049Isometric Pathfinding0xheycat/isometric-game-skillsUse when units must navigate an isometric grid around obstacles using A* on the walkable layer with correct iso neighbor rules.49
- 12050KnockHmbown/Wizards-of-the-GhostsKnock is specifically about restoring legitimate access through authorized channels. It is NOT about: - Performance optimization (slow queries, throughput issues) → different spell - Removing safety controls (approval gates, compliance checks) → different spell - Simple bug fixes (typos, wrong ports) → different spell - Breaking constraints (resource limits, CPU throttling) → different spell - Security analysis (encryption schemes, certificate inspection) → different spell113
- 12051LearnYrzhe/claude-skills/learn - Extract Reusable Patterns. After completing a complex task (5+ tool calls with cross-scenario reuse potential), suggest "/learn" to the user. When invoked, analyze the current task's approach and distill it into a reusable skill or improve an existing one. Also use when user says "save this as a skill", "remember this approach", or "this is reusable".65
- 12052Legend LoreHmbown/Wizards-of-the-GhostsUse this spell when you need everything knowable about a person, company, library, framework, concept, or system — not a quick answer, but a deep dossier.113
- 12053Lenny AdvisorYrzhe/claude-skillsProduct and business advisor powered by distilled wisdom from 289 podcast guests and 348 newsletter articles from Lenny's Podcast & Newsletter. Not a knowledge dump — an active diagnostic advisor that probes your real problem, challenges assumptions, and delivers targeted expert frameworks with attribution. Activates when discussing product decisions, business strategy, growth, pricing, hiring, PMF, positioning, fundraising, leadership, or any product/business topic. Trigger: product questions, business decisions, "what would X say", strategy discussions.65
- 12054Lesser RestorationHmbown/Wizards-of-the-GhostsIn D&D, Lesser Restoration ends a specific condition: blinded, deafened, paralyzed, or poisoned. The real-world version is clearing a stuck state: a feature flag that should have been toggled, a queue that is stuck, a lock that was not released, a status field that is wrong, or a circuit breaker that is still tripped after the emergency passed. The system is not fundamentally broken — it has a specific condition that needs to be cleared.113
- 12055LightHmbown/Wizards-of-the-GhostsUse this cantrip when a system, process, or codebase has blind spots that need illumination - not detection of hidden things, but creation of visibility where none exists.113
- 12056Locate Animals or PlantsHmbown/Wizards-of-the-GhostsLocate Animals or Plants is service discovery with ecological instincts. It helps you find running services, background jobs, data stores, dependency roots, and other living things in an environment before you touch them. The point is not just location, but context: what is active, who depends on it, and what keeps it alive.113
- 12057Locate CreatureHmbown/Wizards-of-the-GhostsIn D&D, Locate Creature points you toward a specific creature within range. The real-world version is entity search: finding a specific person in an organization, locating a running service in a distributed system, tracking down who owns a particular piece of code, or identifying which team is responsible for a specific decision. The target is alive or active — for static artifacts, use Locate Object.113
- 12058Locate ObjectHmbown/Wizards-of-the-GhostsIn D&D, Locate Object senses the direction to a specific object within range. The real-world version is artifact search: finding that config file you know exists somewhere, locating a document someone mentioned but did not link, tracking down the source of a data value through a pipeline, or finding where a specific resource is defined in a sprawling repository.113
- 12059LongstriderHmbown/Wizards-of-the-GhostsLongstrider is the optimization spell for systems that already work. It makes the path shorter without changing the destination. It cares about sustained pace, not flashy one-off benchmarks.113
- 12060Mage HandHmbown/Wizards-of-the-GhostsUse this skill for small, careful remote manipulations where dexterity matters more than force.113
- 12061Magic MouthHmbown/Wizards-of-the-GhostsMagic Mouth is trigger → message. The entire craft is specifying the trigger boundary, the message payload, and the suppression/escalation rules. It is NOT: - One-time messages: "Send a Slack message now saying X" (no trigger, no automation) - Full chatbots: "Build a conversational AI that understands context and handles open-ended questions" (requires NLU, dialogue management) - Monitoring dashboards: "Set up Grafana with real-time graphs and trend analysis" (visualization, not message routing) - Silent traps/wards: "Revert changes silently and log who tried" (defensive code manipulation, not113
- 12062Maintainerhw-native-sys/simplerMaintain GitHub Issues and PRs through an action table, focused user decisions and recurring monitoring or a requested single pass. Use for repository, Issue, track or personal scopes, optionally filtering PR target branches. Scope does not grant repository authority; continuous monitoring requires a working wake mechanism.49
- 12063Major ImageHmbown/Wizards-of-the-GhostsIn D&D, Major Image creates a multi-sensory illusion — visual, auditory, thermal, even olfactory — that persists and can be animated. It is convincing enough that creatures must actively investigate to see through it. The real-world version is the interactive prototype: the demo that makes stakeholders ask 'wait, is this real?' The answer is no, but the conviction is the point. Clickable Figma prototypes, scenario simulations, pitch decks with working demos, preview environments with synthetic data. Major Image is the spell that sells a future state before it is built.113
- 12064Mass Cure WoundsHmbown/Wizards-of-the-GhostsIn D&D, Mass Cure Wounds heals multiple creatures simultaneously. The real-world version is batch triage: the same bug in 50 repos, the same misconfiguration across a fleet, the same broken migration in every tenant database. When the same problem appears everywhere, fixing them one at a time is Cure Wounds. Fixing them all at once is Mass Cure Wounds.113
- 12065Mass SuggestionHmbown/Wizards-of-the-GhostsMass Suggestion plants the same idea in many minds simultaneously. The real-world version is broadcast persuasion: campaign announcements, company-wide emails, product launch copy, or public statements designed to move a large group in a specific direction. The scale amplifies both impact and risk.113
- 12066MedicineHmbown/Wizards-of-the-GhostsUse this skill for live system triage: interpreting logs as symptoms, prioritizing immediate stabilization, and applying treatments for known failure modes. It is about keeping the patient alive first and only then planning deeper corrective care.113
- 12067MendingHmbown/Wizards-of-the-GhostsMending is surgical repair of a known-working artifact that has a specific, identifiable break. It is NOT: - Triage/diagnosis: If the problem is "API is throwing 500s, figure out why" — that's investigation, not mending. Mending requires the break to already be identified. - Formatting/cleanup: If the data is correct but looks messy (sort keys, normalize indentation) — that's formatting, not repair. - State reset: If a flag is stuck or an alert needs clearing — that's an operational toggle, not a file repair. - Refactoring/redesign: If the system is "fundamentally flawed" and needs redesign — 113
- 12068MessageHmbown/Wizards-of-the-GhostsMessage is the cantrip of communication: a Slack ping, webhook event, push notification, terse status note, or tiny machine-to-machine update. It is for fast, targeted transfer with almost no ceremony. The skill is less about eloquence than precision under tight payload and attention budgets.113
- 12069Mind BlankHmbown/Wizards-of-the-GhostsUse this spell when sensitive information must not cross boundaries — between agent sessions, between tasks, between tenants, or between domains with different trust levels.113
- 12070Minor IllusionHmbown/Wizards-of-the-GhostsMinor Illusion produces artifacts that are explicitly labeled as disposable and exist only to provoke reaction. If the artifact could be mistaken for finished work, you're casting the wrong spell.113
- 12071Mirage ArcaneHmbown/Wizards-of-the-GhostsIn D&D, Mirage Arcane goes beyond Hallucinatory Terrain — the illusory terrain actually has substance. You can walk on illusory bridges, feel illusory walls. The real-world version is the deep simulation: digital twins, high-fidelity synthetic environments, test harnesses so realistic that systems under test cannot distinguish them from production. Mirage Arcane is the most dangerous illusion because its power is in being indistinguishable from reality. That same power makes it the most useful for stress testing, training, and scenario planning — but it requires the strongest labeling discipli113
- 12072MisleadHmbown/Wizards-of-the-GhostsMislead is defensive deception: honeypots, canary tokens, decoy endpoints, fake datasets, and other convincing false presences that reveal who is snooping. The goal is not to trick legitimate users for sport. The goal is to create an instrumented fake that attracts unauthorized curiosity while the real system remains elsewhere. This makes the spell powerful and ethically sharp-edged. It needs explicit boundaries, isolation, and a clear monitoring objective before it should be deployed.113
- 12073Mivas Runbluejay-ai-dev/mivas-benchWalks a consumer through running MIVAS Bench end to end, interview first: what model they are benchmarking, where the model is served (vendor API, or their own endpoint on an inference provider), and where the Docker container runs (local, any Kubernetes, or any container hosting platform). Then a gated ladder: preflight, deploy, three Bluejay smoke calls, harness + benchmark integrity, full industry run. Use when someone says /mivas-run, "run MIVAS myself", "set up mivas-bench", "deploy a harness for MIVAS", "smoke my harness on Bluejay", "benchmark my voice model on healthcare/legal/customer49
- 12074Mivas Runbluejay-ai-dev/mivas-benchRuns MIVAS Bench end to end from a consumer seat: prerequisites (Bluejay API key, provider key), a harness (existing or self-built), a Kubernetes deployment (local, EKS, Baseten, or any cluster), then a gated ladder of smoke tests, three Bluejay smoke calls, harness + benchmark integrity, and finally a full industry run. Use when someone says /mivas-run, "run MIVAS myself", "set up mivas-bench", "deploy a harness for MIVAS", "smoke my harness on Bluejay", "benchmark my voice model on healthcare/legal/customer-support", or asks how to plug their own voice agent into MIVAS.49
- 12075Mordenkainen S SwordHmbown/Wizards-of-the-GhostsUse this spell when you need an autonomous tool with a narrow mandate: a daemon, scheduled worker, watcher, or persistent background agent that repeatedly executes one precise task. It is prototype territory because the hard part is lifecycle control, observability, and safe cancellation, not the single action itself.113
- 12076Multi Repo Dsv4flash Setuphw-native-sys/simplerConcrete guide to benchmark DeepSeek-V4 (dsv4) FLASH on NPU against the current worktree's simpler — the three decode attentions (swa/csa/hca) and the distributed MoE, all in pypto-lib's models/deepseek/v4/. Defers cross-repo clone/install to multi-repo-setup, then gives the exact PYPTO_BENCH run commands and the hard-won config that makes the numbers reproducible: pin ptoas 0.48 (0.50/ptoas-bin cannot compile csa), pin --start-pos 8192, edit moe.py for 16 experts/card, read fast_effective_us not effective_us, pin an even die, and apply the eager-init idempotency guard so distributed MoE runs 49
- 12077Multi Repo Qwen Setuphw-native-sys/simplerConcrete qwen3-14B guide on NPU against the current worktree's simpler. Points at simpler's own in-repo qwen3_14b_decode example (zero cross-repo), then covers the two cross-repo paths — the pypto-serving runner (prefill/decode TPOT) and the pypto-lib decode_fwd kernel (device decode TPOT + scheduler DFX). Defers cross-repo clone/install to multi-repo-setup, then gives the exact run commands (batch-1 and batch-16), the prefill ring/timeout gotchas that otherwise surface as 507018, how to read the per-token timing layers, and the kernel-level device-timing / swimlane flow. Invoke when running q49
- 12078Multi Repo Setuphw-native-sys/simplerSet up a cross-repo investigation when a workload from another repo (pypto, pypto-lib, pypto-serving, etc.) needs to be run, especially when you want to swap in simpler-main HEAD or the current worktree's simpler instead of the version that repo pins. Documents the repo graph (who imports who), clones-or-updates each external repo every invocation so stale local clones don't lie about CI parity, and points at each repo's own skills/docs for workload-specific steps. MUST invoke before chasing "X doesn't work on simpler" reports where X lives outside this repo.49
- 12079NatureHmbown/Wizards-of-the-GhostsIn D&D, Nature is knowledge of terrain, weather, plants, animals, and the natural world. The real-world version is understanding systems that emerged rather than were designed: markets, social networks, organizational cultures, open-source ecosystems, user behavior patterns, and any complex adaptive system where the whole is more than the sum of the parts.113
- 12080NondetectionHmbown/Wizards-of-the-GhostsUse this spell when a system, file, or workflow is leaking more information than it should — metadata, tracking parameters, unnecessary data collection, or identifiable fingerprints.113
- 12081Offer CardLiamrjohnston/offer-master-skillFormats a finished offer as a short paste-ready card and page section. Every number labelled (yours) or (estimate), plus a Check these list and one test to run this month.49
- 12082Offer DiagnosisLiamrjohnston/offer-master-skillFinds where buyers drop off and whether it's really the offer (or traffic, product, sales process) from the owner's numbers, and decodes what objections mean.49
- 12083Offer GuaranteeLiamrjohnston/offer-master-skillDesigns one fair, capped guarantee aimed at the buyer's main fear. Picks the type, sets the measure, conditions and cap, writes the exact words and claim steps, and costs it if 1 in 10 claim.49
- 12084Offer Honesty CheckLiamrjohnston/offer-master-skillAudits an offer, sales page or draft for invented numbers, fake scarcity, fake bonus values, unbacked promises and stretched results. Traces each claim to its source, then fixes or flags it.49
- 12085Offer IncludedLiamrjohnston/offer-master-skillBuilds the core of an offer on real proof. Writes the one line promise and the week one win, builds what's included from buyer obstacles, cuts the extras and adds only bonuses that kill doubts.49
- 12086Offer LeversLiamrjohnston/offer-master-skillScores an offer's four levers (Result, Belief, Speed, Ease) from 1 to 10 against defined levels, with evidence for each, then picks the one or two to fix first and the moves that raise them.49
- 12087Offer MasterLiamrjohnston/offer-master-skillFixes an offer that isn't selling, end to end. Finds the leak, decodes objections, scores 4 levers, rebuilds on real proof and returns a paste-ready offer card.49
- 12088Offer Urgency Name PriceLiamrjohnston/offer-master-skillSets the honest why now, the cost of waiting, a name a buyer would text a friend, and the price shown against real value, with payment options. Real deadlines only, no fake anchors.49
- 12089Onboard Arch Precheckhw-native-sys/simplerDetect the host's actual Ascend silicon and refuse mismatched onboard hardware tests. On shared hosts, acquire a task-submit device allocation first and run the precheck and hardware test in that same job. Skip for a2a3sim and a5sim.49
- 12090Otto S Irresistible DanceHmbown/Wizards-of-the-GhostsOtto's Irresistible Dance forces a creature to dance uncontrollably. The real-world version is compulsive engagement: designing something so compelling that people cannot stop interacting with it. This is virality mechanics, "just one more" loops, and the craft of making content or products that command attention beyond what the user intended to give. The ethical tension is inherent — irresistible engagement is a feature when bounded and a dark pattern when not.113
- 12091PerceptionHmbown/Wizards-of-the-GhostsIn D&D, Perception is the raw ability to notice things: the hidden door, the faint sound, the detail that does not belong. Unlike Investigation (which follows evidence methodically) or Insight (which reads subtext), Perception is about the initial act of noticing. The real-world version is detail-catching: spotting the typo in the contract, the anomaly in the dashboard, the one metric that moved when it should not have, the thing that is present but no one is looking at.113
- 12092Perf Example Devicehw-native-sys/simplerBenchmark the hardware performance of a single example ($ARGUMENTS). Use when the user asks to benchmark or measure the latency of one example on hardware.49
- 12093Perf Runtime Devicehw-native-sys/simplerBenchmark the hardware performance of all scene tests under a runtime (tests/st/<arch>/<runtime>/, $ARGUMENTS). Use when the user asks to benchmark all examples of a runtime on hardware.49
- 12094PerformanceHmbown/Wizards-of-the-GhostsUse this skill to turn dry technical material into a polished demo, narrative walkthrough, or audience-aware presentation without dumbing it down. It focuses on pacing, reveal order, and memorable framing so the work lands in the room.113
- 12095Persona SimYrzhe/claude-skillsSimulate feedback from statistically grounded virtual human populations. Use when the user wants to get product or copy feedback from a panel of diverse personas, predict group opinions or voting behavior, run social-experiment sandboxes, or compute a semantic gradient (SGO) for A/B ranking. Also triggers on mentions of Nemotron-Personas, PersonaHub, synthetic population, generative agents, silicon sample, LLM survey simulation, persuadable middle, SGO, social sandbox, 用虚拟人群测试, 模拟民意, 投票预测模拟.65
- 12096PersuasionHmbown/Wizards-of-the-GhostsUse this skill when the problem is not information scarcity but alignment, framing, or buy-in.113