「sam」の検索結果

Designs adaptive clinical trials including group-sequential (O'Brien-Fleming, Pocock, Lan-DeMets spending), sample-size re-estimation (blinded Friede-Kieser, unblinded Cui-Hung-Wang, Mehta-Pocock promising zone), seamless Phase 2/3 with treatment-arm selection, population enrichment, and response-adaptive randomisation. Covers FDA 2019 Final Adaptive Designs Guidance, FDA 2022 Master Protocols, and ICH E20 Step 2b/3 draft (June 2025, NOT final). Use when planning interim analyses, sample-size re-estimation, or master/platform-trial designs.

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。

Persistent compounding memory for AI agents. 5 default MCP tools: session_start, session_end, remember, recall, check. Full surface (18 tools) available with --full flag. Two-verb model: inhale (session_start) and exhale (session_end). Correction-first memory with decision trail tracking, watch_for warnings, palace rooms with salience scoring, cross-project insight matching, same-day journal merging, ambient recall hooks. Local markdown only. Zero cloud, zero telemetry, Obsidian-compatible. Optional Supabase backend: when configured via `ar setup supabase`, recall() uses pgvector cosine similarity on OpenAI/Voyage embeddings instead of keyword search — same API, semantic understanding. Gracefully degrades to local search if not configured.

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。

Plans a controlled experiment (A/B test) so its result can be trusted: writes the hypothesis, picks one primary metric and the guardrails, computes sample size and run time, specifies how visitors are assigned and when exposure is logged, and reads out the result with a confidence interval. Use when someone says "set up an A/B test", "split test this page", "how many visitors do I need", "how long should the experiment run", "is this result significant", "can I stop the test early", or wants to test a headline, price, layout or onboarding change against the current version.

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 a paper's EVALUATION DESIGN actually measures what it claims and whether its reporting is complete — the validity layer family D (experiment-forensics) cannot reach. Three patterns: train/test leakage means the reported score may not measure generalization (HP-EVAL-LEAKAGE — adopts the Kapoor & Narayanan 8-type / 3-category leakage taxonomy; the illegitimate-proxy / sampling-bias / pretraining-contamination subtypes hand off as needs_external_check, naming but NEVER running Oren-2023 exchangeability / Shi-2023 Min-K% / Golchin-2023 Time-Travel / BIG-bench canary); a load-bearing LLM judge is conflicted (same model/family as a compared system) or unvalidated (no human-agreement, no bias control) (HP-JUDGE-VALIDITY); a declared condition/metric is dropped or switched to favor the method, or 'best' is chosen with no held-out set (HP-SELECTIVE-REPORTING). Verdict-bearing at L0/L1 from the DESCRIBED protocol — NOT repo-gated like experiment-forensics; L2 only CONFIRMS against split/preprocessing/result files. A fresh cross-model reviewer (gpt-5.6-sol xhigh, read-only, fresh thread per pass) PROPOSES findings, each span-anchored to a ledger claim_id; tools/adjudicate_findings.py DECIDES the verdict. Leakage and under-reporting are usually HONEST methodological errors — every finding describes a discrepancy to CHECK, never an accusation. An LLM generating GROUND-TRUTH labels is HP-FAKE-GT (experiment-forensics) — routed there, not here. Emits eval-design-forensics.findings.json; computes NO verdict. Detect-only. Triggers: "eval design audit", "evaluation validity", "train/test leakage", "data leakage", "is the score measuring generalization", "LLM judge bias", "is the judge validated", "selective reporting", "cherry-picked results", "评估设计审计", "评测有效性", "数据泄漏", "训练测试集泄漏", "裁判模型有没有验证", "选择性报告".

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。

产品营销 (PMM, Product Marketing) — 拥有「产品如何被定位 / 表达 / 发布 / 卖进市场」的从业者认知操作系统: 是产品 / 销售 / 市场 / 客户之间的连接组织, 负责把产品能力翻译成市场价值, 并把市场声音带回产品。覆盖 (a) 定位与差异化 (April Dunford 「刻意选择语境」框架 + 竞争性替代方案 + 卖价值不卖功能 + 选择「让你优势成立的市场参照系」, 对照 Ries & Trout 经典「定位是占领心智」), (b) 信息传递与叙事 (messaging 层级 / 信息屋 + 价值主张 + 收益导向 vs 战略叙事「换框架」Andy Raskin + 表达清晰度 Emma Stratton / Punchy + 初创 homepage 定位 FletchPMM), (c) 进入市场 (GTM) 策略与产品发布 (发布分级 T1/T2/T3 + 发布流程与跨职能编排 + GTM 打法 PLG vs SLG vs 混合 + 滩头细分选择), (d) 市场与竞争情报 (竞品分析 + battlecard 战卡 + 输赢分析 win/loss + 竞争赋能 Klue/Crayon/Clozd), (e) 买家与客户研究 (ICP 理想客户画像 + 买家 / 用户 persona + Jobs-to-be-Done 任务理论 Christensen/Moesta/Ulwick ODI + 客户之声 VoC + 用真实买家测信息 Wynter/Peep Laja), (f) 销售赋能 (战卡 + pitch / 销售 deck + 一页纸 + demo 叙事 + 异议处理 + 销售培训 + 内容采用率 — 你的产出只值销售实际用到的那部分), (g) 定价与打包 (价值定价 + 打包分层 + 《Monetizing Innovation》Ramanujam/Simon-Kucher + PLG 定价 — 常与产品 / 财务共担), (h) 细分与品类 (市场细分 + TAM/SAM/SOM + 技术采用生命周期与跨越鸿沟 Geoffrey Moore + 品类设计与叙事 Play Bigger / Christopher Lochhead — 稀有且昂贵, 不是默认选项), (i) 需求生成与内容协同 (campaign 信息 + 思想领导力 + 漏斗内容 TOFU/MOFU/BOFU + ABM 基于客户营销), (j) 分析师与影响者关系 (Gartner 魔力象限 + Forrester Wave + 分析师 briefing — To B 场景, 对 pay-to-play 影响诚实标注), (k) 客户营销与倡导 (参考客户 + case study + 评测站 G2/TrustRadius/Capterra 运营 + 社区), (l) PMM 运营与度量 (影响管道 + 赢率 + 发布采用率 + 信息穿透率 + 角色汇报线「向产品还是向市场」之争 + PMM 作为战略职能 vs 接单工具人陷阱); 跨 B2B SaaS (该学科重心) / B2C 消费 / 开发者 PLG / 平台产品。不含 产品管理 (PM 造产品 / 拥有 roadmap, PMM 拥有市场 — 相邻且极易混淆) / 增长 / 效果营销 / 付费投放 (平行学科) / 品牌 / 企业传播 / 公关为终点 / 需求生成为终点 (PMM 与之协同但不等同) / 平面设计 /「做 slide 和周边的团队」(PMM 抗争的接单工具人窄化) / 泛泛「市场营销」。 (Product Marketing (PMM) — the cognitive operating system of practitioners who own how a product is positioned, messaged, launched, and sold into a market: the connective tissue between product, sales, marketing, and customers, responsible for translating product capability into market value and bringing the voice of the market back into product. Covers (a) positioning & differentiation (April Dunford's deliberate-context framing, competitive alternatives, value-not-features, framing the market in which your strengths matter — vs Ries & Trout classic perception-in-the-mind positioning), (b) messaging & narrative (messaging hierarchy / house, value proposition, benefit-led vs strategic-narrative / change-the-frame messaging Andy Raskin, clarity craft Emma Stratton/Punchy, homepage & startup positioning FletchPMM), (c) go-to-market (GTM) strategy & product launches (launch tiers T1/T2/T3, launch process & cross-functional orchestration, GTM motion PLG vs SLG vs hybrid, beachhead segment selection), (d) market & competitive intelligence (competitive analysis, battlecards, win/loss analysis, competitive enablement Klue/Crayon/Clozd), (e) buyer & customer research (ICP ideal customer profile, buyer & user personas, Jobs-to-be-Done Christensen/Moesta/Ulwick ODI, voice of customer, message testing with real buyers Wynter/Peep Laja), (f) sales enablement (battlecards, pitch & sales decks, one-pagers, demo narratives, objection handling, sales training, content adoption — output is only as good as what sales actually uses), (g) pricing & packaging (value-based pricing, packaging & tiering, Monetizing Innovation Ramanujam/Simon-Kucher, PLG pricing — often shared with product/finance), (h) segmentation & category (market segmentation, TAM/SAM/SOM, technology adoption lifecycle & crossing the chasm Geoffrey Moore, category design & narrative Play Bigger / Christopher Lochhead — rare and expensive, not a default), (i) demand-gen & content partnership (campaign messaging, thought leadership, funnel content TOFU/MOFU/BOFU, ABM account-based marketing), (j) analyst & influencer relations (Gartner Magic Quadrant, Forrester Wave, analyst briefings — for B2B, with honest read on pay-to-play perception), (k) customer marketing & advocacy (references, case studies, reviews G2/TrustRadius/Capterra presence, community), (l) PMM operations & metrics (influenced pipeline, win rate, launch adoption, message pull-through, the role's reporting line product-vs-mark

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。

DEPRECATED - Use browsing-bluesky skill instead. Sample and analyze Bluesky firehose to identify trending topics and content clusters. Use when user asks about "what's happening on Bluesky", "Bluesky trends", "zeitgeist", "firehose analysis", or wants to see real-time topic clusters from the network.

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。

Browse Bluesky content via API and firehose - search posts, fetch user activity, sample trending topics, read feeds and lists, analyze and categorize accounts. Supports authenticated access for personalized feeds. Use for Bluesky research, user monitoring, trend analysis, feed reading, firehose sampling, account categorization.

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。

Runs a full dead-end page audit for any website. Discovers all blog/content pages, detects which ones have zero outgoing links to any URL on the same domain, fetches keywords, clusters pages by topic, and generates 3 outgoing link suggestions per dead-end page with anchor text, placement guidance, and ready-to-paste context copy. Outputs a styled HTML report matching report-style-reference.html. Automatically detects site framework (Next.js, static, WordPress) and uses the correct detection method: RSC header for Next.js/Vercel sites, standard curl for static and traditional CMS sites. Trigger when a user provides a domain or URL and asks for a dead-end page audit, outgoing internal links audit, "which pages don't link out", "find pages with no internal links in the content", "pages that are link dead-ends", or any variation of finding pages that have zero outgoing links to the same domain. Always use this skill for dead-end page audits — never attempt the workflow without following every step here.

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。

Suede Labs skill that finds and organizes the rights gaps in a creator project before packaging: ownership, contributors, splits, samples, licenses, provenance, metadata, licensing readiness, and royalty-routing readiness, each marked confirmed or unknown against an evidence trail. Use when a song, release, or creative project needs a rights check before registry, licensing, sync, or payout discussion; when splits, sample clearance, or chain of title are unclear; or when someone asks whether they have the rights to release, license, or get paid for a work. Organizes evidence only: clears no rights, confirms no ownership, moves no money, writes to no registry. NOT FOR: building the transfer package itself (use suede-rights-passport); linting a release folder's files and metadata (use suede-release-linter); a sync one-sheet or pitch (use suede-sync-packaging).

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 building or changing features, preprocessing, encoders, resampling or train/validation/test splits — keep every fitted transform inside a Pipeline fitted on training folds only, and build time and group features without seeing the future or the same entity twice

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。

Bug-fix routine that fixes the whole class of bug, not just the reported instance. Use for any bug report, failing production behaviour, monitoring alert, audit finding or review finding that does real harm (a smaller review finding goes on the item's list, as ship-check says), and whenever the same kind of bug has been seen before. Reproduces it with a failing test, names the failure class, searches the task's scope for the same pattern (the whole codebase when the user's own prompt asks for the fix, not when it is picked from a task's list), fixes or records every hit, and adds the rule, invariant, helper or check that stops it coming back.

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。

Your team has high autonomy — and standups are vague, weekly meetings drift into status reports, and 1:1s produce nothing but "I'm fine." This playbook fixes all three layers at once. What's inside: • Daily Standup system — 3-field format, good sample, 4 COO principles to keep it honest • Weekly Meeting — strict 25-min agenda, Scrum Master rotation, 0–10 pulse check • 1:1 Framework — 9-dimension quantified pulse, 5-group deep question bank, bilateral feedback structure • Diagnostic layer — "stuck-at" detection: info / authority / collaboration / capability / direction • FAQ for 4 common failure modes (dead standups, status-report meetings, "I'm fine" 1:1s, scaling beyond 10 people) 🇨🇳 你的团队自由度很高——但 standup 越来越糊，例会变成汇报，1:1 问不出真话。这套三层 Motivation 操作手册解决的正是这件事：standup 是信息层，例会是对齐层，1:1 是个人层。三件事做好了，motivation 不是管出来的，是长出来的。 🇯🇵 チームの自律性は高いのに、スタンドアップは曖昧で、週次ミーティングは進捗報告になり、1on1では「大丈夫です」しか返ってこない——このプレイブックは、その3層すべてを同時に解決します。 🇰🇷 팀의 자율성은 높은데 스탠드업은 두루뭉술하고, 주간 회의는 진척 보고가 되어버리고, 1:1에서는 진심을 들을 수 없다면 — 이 플레이북이 세 레이어 모두를 동시에 해결합니다. Triggers: "团队 motivation" | "standup 没人写" | "例会变成汇报" | "1:1 问不出真话" | "高自由度团队管理" | "COO 如何管人" | "team motivation" | "standup best practices" | "1on1 framework" | "weekly meeting template" | "チームモチベーション" | "팀 동기부여" | "AI startup team ops" | "remote team alignment" | "engagement下降"

Drive one AI agent from another through kolu's terminals: spawn a Claude Code / Codex / opencode session in a PTY, prompt it, watch the screen, read the reply, prompt again. PRIMARY PATH: the kolu MCP tools (`lifecycle_create`/`lifecycle_sendInput`, `wait_outputSettled`/ `wait_agentState`, `screen_text`, the `terminals` resource). FALLBACK: the `kolu` CLI — same verbs, same discipline. Triggers on "drive another agent", "agent drives agent", "make Claude drive Codex", or wiring a loop where one coding agent supervises another.

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。

A/B test design — produce an experiment spec with hypothesis, primary metric, MDE, sample size, run time, and decision rule. Also determines when NOT to A/B test and what to do instead. Use when asked to "design an A/B test", "should we test this", "experiment design", "how do we know if this works", "what's the sample size", or "set up an experiment".

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。

Foundations of digital signal processing — sampling, aliasing, the Nyquist criterion, quantization, convolution, the discrete Fourier transform, FFT, FIR and IIR filter design, windowing, spectral leakage, and practical issues in fixed-point versus floating-point implementation. Use when designing filters, analyzing spectra, choosing sample rates, reasoning about aliasing, or implementing DSP algorithms on a microcontroller, DSP chip, or host CPU.

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。

A/B testing, randomization, sample size calculation, confounding control, and causal inference for data science. Covers the full experimental lifecycle from hypothesis formulation through power analysis, randomization strategies, blocking, factorial designs, sequential testing, and the potential outcomes framework for causal claims. Use when designing experiments, planning A/B tests, calculating sample sizes, or reasoning about causation from data.

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。

Pick among candidate outputs (code, configs, plans) by running them on diverse inputs and clustering by behavioural fingerprint, rather than by textual aggregation or log-probability. Activates when an executor returns multiple plausible candidates that need disambiguation, when output-majority voting would be the default choice, or when reviewing generated code that has not yet been validated. The 2026 evidence (Semantic Voting, arxiv 2605.08680v1) is that any execution-based selector dominates output-majority voting by 19-52pp; sketch-generated inputs beat random fuzz by 11.3pp. Triggers: "pick the best candidate", "majority vote on code", "select from N samples", "validate the generated output", "behavioural verification".

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。

Cascade the same edit pattern across N files safely. Use when applying the same refactor to multiple files (e.g. swap import paths across 11 scripts, rename a symbol, migrate a call signature). Detects the common-shape-across-files situation and turns an N-file cascade into a planned audit → apply → verify workflow instead of N sequential manual edits.

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。

A **samovar** (самовар) keeps water hot indefinitely through self-contained combustion.

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。

Retroductive hypothesis sampling via MCMC: given observations, importance-sample explanations weighted by P(observation|hypothesis) using Gay.jl colored particles as hypothesis candidates.

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 creating, grooming, planning, or closing a backlog item. Bridges backlog items to SAM planning with issue, project, and milestone tracking against the configured backend. Activates on interactive browsing with no arguments, loading an item by issue reference or title match to run grooming and SAM planning, autonomous unattended runs that substitute evidence-derived decisions for clarifying questions, a quick path for one-file fixes where full grooming is disproportionate, dismissing an item without completion under a required reason (duplicate, out of scope, superseded, wontfix, blocked), marking an item done with an evidence trail and required summary, or initializing GitHub tracking infrastructure for a project. Stops when the item already has a plan or is blocked on missing information.

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。

Routes a Superpowers plan file through the /work-backlog-item pipeline and writes SAM task back-references and chunk annotations into the original plan. Use when given a path to a Superpowers plan file via $ARGUMENTS and needing to create a linked backlog item plus SAM task plan.

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。

SAM-style feature initiation workflow — discovery through codebase analysis, architecture spec, task decomposition, validation, and context manifest. Use when a user asks to add a feature, plan a feature, or convert an idea into an executable SAM plan.

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 the plan_ref returned by add-new-feature is provided. Executes the SAM implementation loop — opens an attempt per ready task, dispatches each to a specialist agent in parallel, settles and judges what comes back, manages bookend tasks (T0 baseline capture and TN verification), and tracks concerns and contract violations per task. Drives task state through the work ledger via the SAM CLI.

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。