5,941 件 ・ 関連度順
概要と使いどころ
Explain what a piece of code does — a specific file, class, or method in close detail, or a user-facing flow as a concise system overview. What it does and why, not whether it's good.
日本語の概要は準備中です。原文の説明を表示しています。
Multi-source research on a customer question or topic with source attribution. Use when a customer asks something you need to look up, investigating whether a bug has been reported before, checking what was previously told to a specific account, or gathering background before drafting a response.
日本語の概要は準備中です。原文の説明を表示しています。
"Apply exponential smoothing methods for time series forecasting with weighted moving averages. Use this skill when the user needs simple, robust forecasts, implement Holt-Winters for seasonal data, or build lightweight forecasting without complex models — even if they say 'simple forecast', 'moving average prediction', or 'smoothing method'.".
日本語の概要は準備中です。原文の説明を表示しています。
Investigation flow (Glob -> Grep -> Read), evidence-based research with file:line references, structured output format for AI consumption. Use for pattern discovery, implementation research, and codebase investigation.
日本語の概要は準備中です。原文の説明を表示しています。
Together AI SDK patterns for TypeScript — client setup, chat completions, streaming, structured output, function calling, embeddings, image generation, fine-tuning, and OpenAI-compatible endpoints
日本語の概要は準備中です。原文の説明を表示しています。
Reconnaissance and methodology playbook. Use when mapping assets, discovering endpoints, fingerprinting technology, and building a structured testing plan for a new target.
日本語の概要は準備中です。原文の説明を表示しています。
This repo is indexed by graft/. For ANY task here, whether understanding how something works, finding where code lives, tracing what calls a symbol or what a change breaks, or scoping an edit, get your context from graft before grepping or reading source files.
日本語の概要は準備中です。原文の説明を表示しています。
DNA methylation analysis — array and bisulfite-seq processing, DMP/DMR calling, and epigenetic clocks.
日本語の概要は準備中です。原文の説明を表示しています。
Evaluate AI agents rigorously — task benchmarks, success criteria, failure taxonomy, cost/latency tracking, and regression testing. Use when you need to know whether an agent actually works, not whether it demos well.
日本語の概要は準備中です。原文の説明を表示しています。
Check whether the work is actually finished rather than finished-looking — what was promised versus what really runs. Several models compare the task against the code, and every completion claim has to survive a command that was actually executed. Use before calling something done, before a PR, at the end of a session, or on "is this done", "did I finish", "check if done", "what did we skip".
日本語の概要は準備中です。原文の説明を表示しています。
Post-mortem analysis of a PhysicsIntern workspace run (Claude Code, Pi, Codex, or OpenCode host). Reconstructs the trajectory from the session record — JSONL file(s) for Claude/Pi/Codex, the SQLite store for OpenCode — audits methodology adherence against the workspace's own CLAUDE.md / AGENTS.md and skill/agent prompts, checks commit discipline and flag dispositions, and assesses substantive quality. Produces a thorough evidence-anchored markdown report. Use after a workspace has been worked on to identify what went well, where the methodology slipped, and what prompts to improve.
日本語の概要は準備中です。原文の説明を表示しています。
Product marketing and go-to-market strategy for B2B SaaS launches and positioning. Use this skill when planning a product launch, developing positioning and messaging, analyzing competitive landscape, building customer advocacy programs, designing a customer referral program (who may accept the reward, how the introduction is made), designing a win/loss interview program, working out why we lose deals, designing GTM strategy, creating category positioning, planning analyst briefings, or deciding which country to expand into and how far to localize. Also use it for an early-stage startup where the founder is still doing the selling — running founder-led sales, setting up a design-partner program, judging whether you have product-market fit, deciding what not to do yet, deciding when to hire the first AE or first marketer, and moving upmarket to larger buyers. Also use it when your company has just acquired another (or is about to), to plan the marketing integration: what you may and may not do between signing and close, the first letter to the acquired customers, whether to keep, endorse, fold in or retire the acquired brand, merging the two marketing databases without mailing people who opted out, when to start cross-selling, and announcing a product sunset. Also use it to design a demand-generation campaign and the channel plan it runs on: one campaign with one job, the offer for buyers who are and are not in market, what each channel is for, the kill and scale rules set before launch, the follow-up sales agrees to, the yearly calendar of always-on programs and tentpole campaigns, and which channels to keep, test or stop. Also use it for brand marketing and demand creation — how much to spend on brand versus performance, what to measure when a brand campaign produces no conversion event, and how to answer "we cannot attribute it". Also triggers on: founder-led sales, founder-led GTM, design partners, design partner agreement, first customers, first 10 customers, do we have product-market fit, PMF, Sean Ellis test, very disappointed survey, pre-seed, seed stage, when to hire our first AE, first sales hire, first marketing hire, moving upmarket, going upmarket, enterprise readiness, first enterprise deal, what do enterprise buyers need before they sign, positioning, product launch, beta program, early access program, design partners, private preview, public preview, general availability, GA launch, when can we call it GA, can sales sell a beta feature, selling the roadmap, launch tier, competitive intelligence, messaging, customer advocacy, win/loss analysis, win/loss interviews, why we lose deals, closed-lost analysis, loss reason, close-reason field, competitive displacement, displace an incumbent, rip and replace, competitive takeout, switch from a competitor, win deals from a competitor, incumbent contract, renewal window, auto-renewal notice period, switching costs, EU Data Act, cloud switching rights, why do we lose to no decision, GTM, category design, ideal customer profile, ICP, what is our real ICP, who actually buys from us, refresh our ICP from closed-won deals, ICP refresh before planning, analyst briefing, G2 review, international expansion, market entry, expand into Europe, which country should we launch in next, localization, translate our site, in-country go-to-market, regional GTM, is cold email legal in Germany, brand marketing, brand awareness, brand vs performance, brand versus demand gen, demand creation, demand capture, 95-5 rule, out of market buyers, category entry points, distinctive brand assets, mental availability, brand tracking, brand health, brand lift, share of search, how do we measure brand, justify brand spend, brand budget. Also triggers on public-sector marketing: sell to government, government marketing, public sector go-to-market, federal marketing, state and local government, SLED, FedRAMP, FedRAMP marketing claims, can we say FedRAMP ready, GovRAMP, StateRAMP, VPAT, accessibility confor
日本語の概要は準備中です。原文の説明を表示しています。
Complete methodology for computing publication-quality cumulative abnormal returns with proper event-study test statistics, matching the robustness of Kaspereit's eventstudy2 for Stata. Covers dateline construction, event-date mapping, estimation and event windows, thin-trading adjustment, OLS with Theil prediction error correction, abnormal return computation, CAR/CAAR/AAR accumulation, boundary contamination guards, and common tests such as Patell, BMP, Kolari-Pynnonen, generalized sign, Wilcoxon, and GRANK-T. Use when the user mentions abnormal returns, event windows, market-model regressions, CARs, CAAR, AAR, eventstudy2, thin trading, trade-to-trade returns, or event-study test statistics.
日本語の概要は準備中です。原文の説明を表示しています。
Diagnose whether an LLM classifier's validation-gate failure is GOLD-BOUND before spending on prompt revision or model changes. Use when: (1) a scoring pipeline over-predicts a label (precision low, recall high) and a prompt clarification is proposed to tighten it, (2) a pilot/validation gate fails and the fix candidates are prompt edits, (3) inter-rater agreement on the weak label was already low (κ < ~0.6). Core check: if gold POSITIVES share the exact feature the revision would exclude, no prompt can pass a gold-scored gate — recall craters while precision barely moves. Also documents the verified surgical-pilot design (single-section diff, tune/holdout split, pre-registered gate, perturbation check on untouched sections).
日本語の概要は準備中です。原文の説明を表示しています。
N-round adversarial review pipeline for empirical research output — the chain from data to LaTeX tables to a manuscript that cites them. A Claude drafter proposes minimal diffs, a deterministic mechanical battery gates every diff from a clean state with a regression gate, a Codex reviewer files check-backed critiques, and a blind judge panel decides residual disputes. Manual-invoke ONLY: trigger when the user explicitly runs /adversarial-empirical-review or names 'adversarial-empirical-review' / 'adversarial empirical review'. Do NOT auto-trigger on generic 'review my results', 'check my tables', or manuscript-editing requests. For prose-style refinement use style-emulation instead; this skill AUDITS WHETHER THE TABLES ARE CORRECT — that each number in the tables is what the analysis code computes, reproduces from the data, and is internally consistent. It is an empirical + code review: the manuscript is read only to resolve table numbering, and prose is not examined.
日本語の概要は準備中です。原文の説明を表示しています。
Audit whether all score-deciding logic in sgl-eval is vendored from NeMo-Skills, or whether some has crept into SE code. Use when the user asks "are we vendoring enough", "review vendor coverage", "audit vendoring", "is our vendoring complete", or before a release.
日本語の概要は準備中です。原文の説明を表示しています。
Use when a task needs the judgment of a Drilling and Boring Machine Tool Setter, Operator, or Tender — deciding whether a hole's depth-to-diameter ratio requires peck drilling, establishing a straight hole start before wander compounds, reducing cutting parameters for a long-reach boring setup versus a short-reach one, or diagnosing whether a dimensional drift traces to tool wear rather than assuming the tool still looks sharp.
日本語の概要は準備中です。原文の説明を表示しています。
Use when a task needs the judgment of a Driver/Sales Worker on a direct-store-delivery (DSD) route — negotiating a shelf-facing or planogram change with a store manager, resizing a standing order after a shelf or demand change, deciding whether to pull code-dated product before a stale credit hits, reconciling a route's revenue-per-stop against cost-per-stop, or deciding whether to fight for shelf space back from a competitor.
日本語の概要は準備中です。原文の説明を表示しています。
Use when a task needs the judgment of an Epidemiologist — investigating a disease outbreak and identifying its source, designing a case-control or cohort study to test an exposure hypothesis, calculating attack rates/odds ratios/relative risk from surveillance data, evaluating whether a surveillance system is sensitive and timely enough to catch a real signal, or judging whether an observed case cluster is a true outbreak or expected background variation.
日本語の概要は準備中です。原文の説明を表示しています。
Use when a task needs senior data-scientist judgment — designing or reading out an experiment, investigating a metric move, deciding whether a causal claim holds, or judging whether a model or analysis actually supports the decision it's being used for.
日本語の概要は準備中です。原文の説明を表示しています。
Use when a task needs the judgment of a detective/criminal investigator — developing and stress-testing a case theory against alternate explanations, reconciling conflicting timestamps (DVR, cell tower, witness-reported) into a defensible timeline, structuring a suspect/witness interview using a non-accusatorial method, drafting a probable-cause affidavit, or evaluating whether a confession or eyewitness identification is being treated with the right level of certainty.
日本語の概要は準備中です。原文の説明を表示しています。
Use when a task needs the judgment of a cytotechnologist — screening a gynecologic (Pap test) slide and applying Bethesda System criteria, assessing non-gynecologic specimen adequacy (FNA rapid on-site evaluation, body fluid, brushing) against the relevant reporting system, deciding whether a finding can be signed out independently or must be referred to a pathologist, or auditing screening quality metrics (ASC:SIL ratio, rescreen concordance, workload compliance). A reasoning aid, not a substitute for a certified cytotechnologist or pathologist signing out an actual case.
日本語の概要は準備中です。原文の説明を表示しています。
Use when a task needs the judgment of a Computer User Support Specialist — triaging an inbound ticket's severity and SLA clock before troubleshooting it, verifying a caller's identity before any password/MFA/account-state change, walking a live incident through a structured troubleshooting sequence instead of guessing at a fix, deciding whether a request is an incident, a service request, or the underlying problem behind a string of incidents, or judging whether a ticket should resolve at tier 1 or genuinely needs escalation.
日本語の概要は準備中です。原文の説明を表示しています。
Use when a task needs the judgment of a CNC Tool Operator — diagnosing a dimension trending toward its tolerance limit across consecutive parts before it actually crosses, deciding whether to apply a tool wear offset or flag the tool for replacement, evaluating whether a failed part traces to work-holding repeatability versus a program or tool issue, or verifying a setup change with a first-article inspection before running production quantity.
日本語の概要は準備中です。原文の説明を表示しています。