AI & ML
AI & ML agent skills cover the machine-learning workflow itself: writing and evaluating prompts, building RAG pipelines, running evals, and wiring up model APIs. Each one is a SKILL.md file your agent loads on demand, so the know-how travels across Claude Code, Cursor, and 60+ agents.
-
houjingyi00417-hub Skill StatsmodelsStatistical models library for Python. Use when you need specific model classes (OLS, GLM, mixed models, ARIMA) with detailed diagnostics, residuals, and inference. Best for econometrics, time series, rigorous inference with coefficient tables. For guided statistical test selection with APA reporting use statistical-analysis.
-
houjingyi00417-hub Skill Scikit LearnMachine learning in Python with scikit-learn. Use when working with supervised learning (classification, regression), unsupervised learning (clustering, dimensionality reduction), model evaluation, hyperparameter tuning, preprocessing, or building ML pipelines. Provides comprehensive reference documentation for algorithms, preprocessing techniques, pipelines, and best practices.
-
idoforgod Bundle Vision Foresight Futures Wheel Basic V1TLDR — Jerome C. Glenn (2009 [1971], V3.0 06장 §III.A) Basic Futures Wheel V1 sub-skill + 박사님 2026-05-11 6차 강화. **vision-foresight-futures-wheel** 마스터 전용. 중앙 trend/event 주위로 Primary→Secondary→Tertiary→Quaternary→Quinary→Senary 6 동심원 ring을 펼침. 6차 default 강제, 각 ring 진입 직전 deep-reasoning-engine PRRG Gate 자동 호출(blocking). **결정론 연산(입력 검증·WebSearch count·부호역전 검사·SRS 계산·ring count 검증·CoER 검증·ring table 생성)은 wheel_engine.py 위임**; LLM 재계산 금지. ## Triggers — INTERNAL ONLY. 마스터 Cycle 1·7에서 자동 호출. 사용자가 'Basic Futures Wheel', 'V1', '자유형 wheel', 'concentric ring' 명시 시 마스터가 본 sub-skill 발동. ## Detailed Methodology — Glenn (2009) §III.A 풀 구현 + 박사님 2026-05-11 6차 강화. 9 Phase: ① Phase 1 Center Definition — 4요소 명시. ② Phase 2 Primary Ring (T+1~5y) — Gate_P1_Pre 통과 후 5~10개 산출. ③ Phase 3 First Ring Closure. ④ Phase 4 Secondary Ring (T+5~10y) — Gate_P2_Pre 통과 후 2~3 secondary 분기. ⑤ Phase 5 Tertiary Ring (T+10~20y) — Gate_P3_Pre 통과. ⑥ Phase 6 Quaternary Ring (T+15~25y, 세옹지마 1차 반전) — Gate_P4_Pre 통과(SIGN REVERSAL ≥50% 강제). ⑦ Phase 7 Quinar
-
idoforgod Bundle Vision Foresight Wild Cards Impact IndexTLDR — Wild Cards Sub-skill ③ Arlington Impact Index. Petersen & Steinmüller(2009) V3.0 10장 Section III.2 "Impact factors: The Formula for Wild Cards" 풀 구현 **INTERNAL** sub-skill. **공식: ΔC + R + V + O + T + Op + P = I_AI**. 7 factor 풀 적용 — ① Rate of change ΔC(1=years, 2=months, 3=days) ② Reach R(1=local→5=global) ③ Vulnerability V(1=less→3=more) ④ Outcome O(1=less→3=more uncertain) ⑤ Timing T(1=late→4=sooner) ⑥ Opposition Op(-2=much support→+2=much opposition) ⑦ Power Factor P(1=Tools→4=Being). 이론 범위 1~24 (PDF p.20 Figure West Coast Disaster example = 19). 별도 Quality Factor (+positive·-negative·±both) — 직접 산입 X. 별도 Foresight Factor A-F (sources many→few). PDF Section III.3 verbatim "Quality variable gives the prediction of the net effect of each (note that this is not derived directly from the Arlington Index)". AI Arlington Impact Calculator Agent 자동 작동 — 외부 평가단 미동원. ## Triggers — INTERNAL sub-skill. 마스터 vision-foresight-wild-cards가 Cycle C1·C3·C7 발동 시 자동 호출. 사용자 직접 호출 불가 (disable-model-invocation: true). 호출
-
idoforgod Bundle Vision Foresight Scenarios Policy TestingTLDR — INTERNAL Sub-skill. vision-foresight-scenarios 대표 마스터의 10번째 sub-skill. Glenn & TFG V3.0 19장 Section III TFG Reporting and Utilization "Testing policies" + Section II "dichotomize strategies between robust and contingent elements" verbatim 풀 구현. **Policy Tester AI Agent** — PDF 원전 verbatim: *"Testing policies. The range of scenarios can be used to test policies. In any study, a list of alternative actions is prepared. This list may come from the decision makers after reading the scenarios. Each is defined as precisely as possible. Then, using quantitative techniques if possible, the policies are 'tested' in each of the scenarios. When a particular policy produces desirable results in all cases, it is clearly a good bet. The other scenarios may give rise to contingent policies that can be called on if the circumstances develop that the scenarios depict."* Section II 8 활용 verbatim 중: *"dichotomize strategies between robust and contingent elements"*. **Robust strategies** = 모든/대부분 scenarios에서 desirable res
-
idoforgod Bundle Vision Foresight Futures Wheel Temporal V3TLDR — Jerome C. Glenn (V3.0 06장 §VI Frontiers, "Version 3") Temporal V3 sub-skill. **vision-foresight-futures-wheel** 마스터 전용. Glenn 직접 인용: *"A 'Version 3 Futures Wheel' would add the dimension of historic forces, current correlations, and future implications in a cone-like fashion."* Figure 7의 3D cone — 위(Future Consequences) · 중간(Current Impacts) · 아래(Historic Forces) — 3 team 작업(historic team / contemporary team / future team)을 AI Agent 3 Team Panel로 대체. ## Triggers — INTERNAL ONLY. 마스터 Cycle 3(Temporal V3) 또는 Cycle 7(Full Stack)에서 자동 호출. 사용자가 'V3', 'Version 3', '시간축', 'temporal cone', '3D cone', 'historic forces', 'current correlations', 'future implications', '과거 동력 + 현재 영향 + 미래 결과', '역사적 맥락 같이', '3D wheel', 'cone-like' 명시 시 마스터가 본 sub-skill 발동. ## Detailed Methodology — Glenn (2009) §VI "Version 3" + Figure 7 풀 구현. 7 Phase: ① **Phase 1 Center Definition + Temporal Anchor** — T+0 시점 정의 + 과거 lookback 범위(보통 20~50년) + 미래 lookahead 범위(보통 10~30년). ② **Phase 2 Historic Forces Team** — Historic Team Panel(역사학자·
-
idoforgod Bundle Vision Foresight Wild Cards IdentificationTLDR — Wild Cards Sub-skill ① Identification. Petersen & Steinmüller(2009) V3.0 10장 Section III.1 "Identification: What Wild Cards can in principle happen?" 풀 구현 **INTERNAL** sub-skill. 5 method orchestration — ① Brainstorming (positive + negative WC 모두) ② Expert Interviews (AI persona panel) ③ Surveys (Web-based open + seed) ④ Historical Analogies (1918 flu·dotcom·9/11·SARS·2008 meltdown) ⑤ Science Fiction (Ted Chiang·Cixin Liu·Vernor Vinge·Octavia Butler·Stanislaw Lem). Petersen 79-entry catalogue (원서 "78-catalogue" 명칭; master §10 추출본 79항) + Steinmüller 55-catalogue 자동 cross-reference. STEEP categorization (Society·Technology·Economy·Environment·Politics + Spiritual 추가). 3 Surprise Types tagging (Type1 next-earthquake·Type2 climate-impact·Type3 unknown-unknowns). 20-40 candidates default. 외부 사람·전문가 미동원 — 5 AI Agent (Brainstormer·Expert Persona·Survey Aggregator·Historical Detector·SF Scanner) 자동 작동. 결정론 확인: catalogue_lookup.py (P-XXX-NN 코드) + validate_identification.py (Type3≥20%·구조 검증) 의무 실행. ## Triggers —
-
idoforgod Bundle Vision Foresight Wild Cards Options ActionTLDR — Wild Cards Sub-skill ⑤ Options for Action. Petersen & Steinmüller(2009) V3.0 10장 Section III.4 "Options for Action: Is there anything we can do about them?" 풀 구현 **INTERNAL** sub-skill. PDF 핵심 원리 verbatim: "Dealing with Wild Cards requires innovative, unconventional methods." 3축 풀 자동화 — ① **5 Nonlinear/Out-of-the-box Thinking** (Systems thinking·Creativity training·Intuition·Associative thinking·Dreamwork — "All are technologies for shaking up old assumptions and allowing new ideas to emerge") ② **3 Basic Rules** (Rule I think now·Rule II information key·Rule III extraordinary approaches) ③ **9-Step Institutional Process** (high-interest 식별·segment·lesser-events·scouting·awareness·structure·display·decide·action·gates). 8-segment Wild Card 분류 (must·can·only-prepare·no-warning·too-big·can-change·new-solution·existing-tools = S1~S8). Action plan + gate/trip-wire 설계. AI Options Strategist Agent 자동 작동 — 외부 정책 결정자·전략 컨설턴트 미동원. 결정론 환원: 모든 segment/rule/step/conceptual redefinition/outcome polarity 사실 조회는 `opt
-
idoforgod Bundle Vision Foresight Futures Wheel Delphi RoundsTLDR — Jerome C. Glenn (V3.0 06장 §VI Frontiers, "Another future variation of the Futures Wheel could use a Delphi via the Internet") Delphi Wheel sub-skill. **vision-foresight-futures-wheel** 마스터 전용. Glenn 직접 인용: *"An international panel could assemble asynchronously to systematically construct a Futures Wheel"* — 5-round async Delphi via Internet + frequency-weighted oval size + Wiki collab + groupware support. Async panel 5~30인을 AI Agent Async Panelists로 대체. `foresight-expert-pool`·`foresight-delphi`·`foresight-realtime-delphi`와 cross-skill 호환. ## Triggers — INTERNAL ONLY. 마스터 Cycle 6(Delphi Wheel)에서 자동 호출. 사용자가 'Delphi-based wheel', '5-round async wheel', 'Delphi via Internet', 'Wiki collab', 'asynchronous panel', 'geographically dispersed', 'frequency-weighted oval', 'Futures Wheel Wiki', '국제 패널 합성', '비동기 wheel 작성' 명시 시 마스터가 본 sub-skill 발동. ## Detailed Methodology — Glenn (2009) §VI "Frontiers... Delphi via the Internet" 풀 구현. 8 Phase: ① **Phase 1 Panel Recruitment** — AI Agent Async Panelists 5~30인 캐스팅(`
-
meowandy Bundle Skill Astrbot DevReference + workflow notes for AstrBot plugin development (messages, platform adapters, plugin config, agent system, i18n, pages).
-
idoforgod Bundle Vision Four Futures최윤식 박사 『미래준비학교』(2016, 지식노마드, ISBN 9788993322972)의 *4가지 미래 가능성*을 그대로 구현한 비전 미래학 코칭 스킬. 박사님 미래학자 본업의 핵심 분류 — ① **Plausible Future (기본미래)** — 발생할 가능성이 논리적으로 충분한 미래·트렌드+계획+심층원동력+대중 이미지 4요소로 구성 ② **Possible Futures (가능 미래들)** — 기본미래에 풍부한 상상력 더해 폭넓게 확장된 가능성 ③ **Wildcard/Unexpected Future (뜻밖의 미래)** — 가능성 낮으나 특정 조건 충족 시 파괴적 영향력 — 비약적 진보 + 붕괴 후 새로운 미래 ④ **Normative/Preferred Future (바람직한 미래)** — 꿈과 가치가 어우러진 *원하는 미래*. 박사님 미래학 정의 명확화 — *"하나의 미래(a Future)" 또는 "그 미래(the Future)"가 아닌 "대안적 미래들(Alternative Futures)"과 "다양한 가능성의 미래들(Possible Futures)"*. 본 스킬은 **결정론적 파이썬 백본**(facts.json·citations.json·external_sources.json + scripts/lookup.py·render_skeleton.py·validate_output.py)으로 사실 조회·라벨 매핑·인용 출처·출력 골격·검증을 모두 결정론 환원하여 LLM 자연어 추론의 환각을 구조적으로 차단한다. 학술적 위상 외부 출처(Henchey 1978·Hancock & Bezold 1994·Voros 2001/2003·Taylor 1990·Petersen 1997 / Copenhagen Institute 1992)는 모두 web 검증 완료. 사용자가 "4가지 미래", "Plausible Possible Wildcard Normative", "기본미래 가능미래 뜻밖의미래 바람직한미래", "Alternative Futures", "박사님 미래 가능성 분류"를 언급하거나 박사님 비전 5단계 Stage 2(비
-
l3mpire Skill Prompt EngineeringTransforms any rough, vague, or underperforming prompt into a production-ready, optimized prompt for Claude. Use this skill whenever the user wants to improve a prompt, says "make this prompt better", "optimize this", "my prompt isn't working", "write me a prompt for X", or shares any instruction they want Claude to follow reliably. Covers system prompts, user prompts, and agent/workflow prompts. Always produces the optimized prompt + a clear explanation of every choice made.
-
idoforgod Bundle Vision Foresight Scenarios Narrative WritingTLDR — INTERNAL Sub-skill. vision-foresight-scenarios 대표 마스터의 7번째 sub-skill. Glenn & TFG V3.0 19장 Section II + III Schwartz Step 5 + Appendix A·B verbatim 풀 구현. **Narrative Writer AI Agent** — Schwartz Step 5 verbatim: *"fill out the scenarios"*. PDF Section II 핵심 verbatim: *"A scenario is a story with plausible cause and effect links that connects a future condition with the present, while illustrating key decisions, events, and consequences throughout the narrative."* + *"sufficiently vivid so that one can clearly see and comprehend the problems, challenges, and opportunities that such an environment would present"*. TFG Development verbatim: *"Prepare descriptions. Now, given the quantitative forecasts of the measures based on the probabilistic description of the impacting events, many chains of causality become apparent, and cohesive narratives describing the future histories can be prepared."* Section II 추가: *"future history—that is, the evolution from present conditions to one of several futures... lay
-
idoforgod Bundle Vision Foresight Scenarios Projection EngineTLDR — INTERNAL Sub-skill. vision-foresight-scenarios 대표 마스터의 6번째 sub-skill. Glenn & TFG V3.0 19장 Section III TFG Development "Project the key measures" + Section V Frontiers TIA conjunction verbatim 풀 구현. **Projection Engineer AI Agent** — PDF 원전 verbatim: *"Project the key measures. Trend Impact Analysis (TIA) is a useful technique for projecting the key measures. (A methodology paper in this series describes this technique.) Briefly, the historical data for each of the measures is projected using time-series methods. The events, expressed probabilistically, are combined with the extrapolation using Monte Carlo methods to produce a new median forecast and a range of uncertainty. Since events within a scenario impact several measures wherever they are used, they have the same probability; thus, internal consistency is promoted."* Section V: *"Also, over the years, various quantitative forecasting methods, such as trend impact analysis (TIA), have been used in conjunction with scenarios. A late 1980s scenario
-
idoforgod Bundle Vision Foresight Scenarios Leading IndicatorsTLDR — INTERNAL Sub-skill. vision-foresight-scenarios 대표 마스터의 11번째 sub-skill. Glenn & TFG V3.0 19장 Section III Schwartz Step 7 + Section V Frontiers Z_Punkt early indicators verbatim 풀 구현. **Leading Indicators Monitor AI Agent** — Schwartz Step 7 verbatim: *"select the leading indicators and signposts for monitoring purposes"* (Schwartz 1991 pp.226-234). Section IV strength verbatim: *"Instead of each possibility being a potential threat to a rigid plan, they tend to be evaluated as signposts, indicating paths along the way to alternative and anticipated futures."* Z_Punkt Section V verbatim: *"Z_punkt the Foresight Company work with attaching 'early indicators' to scenarios. These lists of indicators, qualitative as well as quantitative, serve as guidelines in monitoring change in the environment, and are regularly checked for possible changes / new trends, for example in a six-month rhythm. Decision makers are thereby enabled to continuously check whether many indicators point into the direction of one or s
-
idoforgod Bundle Vision Foresight Environmental Scanning ReportTLDR — Gordon·Glenn (2009) Appendix B 양식 (Millennium Project Environmental Security 월간 보고서) 풀 구현 INTERNAL SUB-SKILL. ## Triggers — 사용자 직접 trigger 차단 (disable-model-invocation: true). 대표 스킬 Step 6·Cycle C에서 AI Report Editor Agent가 자동 호출. ## Detailed Methodology — 8 Items 양식 (Title·Summary·Implications·Sources). 사용자가 대표 스킬 Step 1-A에서 선택한 N개 Implications 도메인 (정부·기업·학계·미디어·교육·종교·금융·비영리·개인·custom) 자동 mapping. Cross-month Issue ID tracking. Body 8~15p + Appendix 5~20p 완성본.
-
idoforgod Bundle Vision Foresight Scenarios Cone Of PlausibilityTLDR — INTERNAL Sub-skill. vision-foresight-scenarios 대표 마스터의 9번째 sub-skill. Glenn & TFG V3.0 19장 Section V Frontiers Charles W. Taylor *"Cone of Plausibility"* verbatim 풀 구현. **Cone of Plausibility Modeler AI Agent** — PDF 원전 verbatim (Section V): *"Charles W. Taylor, a strategic futurist with the U.S. Army War College, has developed a 'Cone of Plausibility,' a theoretical process that can be used holistically to project trends and events and their consequences into the future and to generate alternative scenarios at predetermined points in time (Chemtech 1993). The Cone of Plausibility encompasses theoretical projections of four planning scenarios; each having a dominant theme or driver, such as technology, politics, economics, and sociology (Exhibit 3). Each of these themes or drivers commands a vision or scenario of the future. Trends within each theme are affected by interactions among trends in which dominant trends alter the less dominate ones or result in discontinuities. Outside of the cone are wild-
-
idoforgod Bundle Vision Foresight Wild Cards Scenario IntegrationTLDR — Wild Cards Sub-skill ⑥ Scenario Integration (Cycle C6·C7). Petersen & Steinmüller(2009) V3.0 10장 Section V "Use with Other Methods" 풀 구현 **INTERNAL** sub-skill. PDF 핵심 verbatim: "Wild Cards are quite effective when used with scenario-building" (Wack 1985). Steinmüller 1997 **6 selection rules** 자동 적용 — ① appropriate to situation ② original and new ③ "barely possible" prioritized ④ analysis not limited to 2-3 ⑤ negative WCs first (test scenario stability) ⑥ contextual + peripheral combined. 4 scenario-WC usage modes (PDF Section V): test susceptibility · compensate weak points in mental map · recognize new alternatives · fight wishful thinking / hyper worst case. ALARM EU 6th Framework Program modeling case 자동 implementation — energy price shocks·contagious natural epidemics·thermohaline collapse North Atlantic cooling as exogenous excitation. AI Scenario Integrator Agent 자동 작동 — 외부 시나리오 워크샵 미동원. ## Triggers — INTERNAL sub-skill. 마스터 vision-foresight-wild-cards가 Cycle C6·C7 발동 시 자동 호출. 사용자 직접 호출 불가 (dis
-
kylin-feng Bundle AI Automation RecipesUse when building AI automation workflows, chaining MCP tools, or creating content production pipelines. Ready-to-use recipes combining multiple MCP servers for end-to-end automation.
-
yu-xuan99 Skill Opencli UsageUse at the start of any OpenCLI session — this is the top-level map of what `opencli` can do, how to discover adapters, what flags and output formats are universal, and which specialized skill to load next. Point here when an agent asks "what can opencli do?" or "how do I find the right command?".
-
andrewnggirl Bundle Skill ScorerEvaluates Agent Skills (Cursor / Claude / OpenClaw compatible) and produces a quantitative, rubric-based score with actionable improvement suggestions. Use when the user asks to review, rate, audit, grade, lint, or improve a SKILL.md file, a skill folder, or a skill archive, or says things like "给这个 skill 打分", "评估一下 skill 质量", "audit this skill", "rate my agent skill".
-
idoforgod Bundle Vision Foresight Environmental Scanning TechniquesTLDR — Gordon·Glenn (2009) Section III의 6 정보 수집 기법 풀 구현 INTERNAL SUB-SKILL. ## Triggers — 사용자 직접 trigger 차단 (disable-model-invocation: true). vision-foresight-environmental-scanning 대표 스킬이 Step 2·Step 4에서 자동 호출. ## Detailed Methodology — 6 기법 (Expert Panels·Database Lit Review·Internet Searches·Hard-Copy·Essays·Key Person Tracking). AI Scanner Agent가 web tools로 자동 수집 + AI Expert Panel 30~75 페르소나 시뮬레이션. 외부 사람 동원 X.
-
idoforgod Bundle Vision Foresight Wild Cards Implications SynthesisTLDR — Wild Cards Sub-skill ⑦ Implications Synthesis. Petersen & Steinmüller(2009) V3.0 10장 통합 합성 + Section II "Extraordinary Implications"·Section IV Strengths/Weaknesses·Section VI Frontiers 풀 구현 **INTERNAL** sub-skill (final stage). PDF 핵심 verbatim: "The study of Wild Cards is particularly important now because the extraordinary growing technological capability of humans has produced, starting a few decades ago, new classes of Wild Cards. For the first time, Wild Cards have global implications. In some cases scientists believe they could potentially threaten the whole human race." 4 핵심 산출 — ① Domain별 implications (Step 0 사용자 선택 도메인) ② 78+55 catalogue cross-reference final mapping ③ Section IV Weakness disclaimer 자동 첨부 ④ Section VI Frontiers options 자동 제안 (Wild Card databases·Internet portals·early warning systems·iKnow·Surprise Anticipation Centers). PDF "Possible Implications" 4-axis Petersen 양식 (REALITY·HABITAT·ACTIVITY·GROUP RELATIONSHIPS) verbatim 적용. AI Senior Wild Card Synthesizer Agent 자동 작동. ## Tri
-
idoforgod Bundle Vision Foresight Futures Wheel Deep Reasoning EngineTLDR — Futures Wheel **추론 검증·강화 엔진** sub-skill. `vision-foresight-futures-wheel` 마스터 전용. 박사님 2026-05-11 강화 명령 3 요구사항 통합 구현: ① 6차 이상 깊이 강제 (MDE) ② 차수마다 pre-research gate (PRRG 6 Gates) ③ 출처·근거 명시 강제 (CCP + CoER). *공상* 차단 + *근거 있는 6차 깊이 추론* 보장. 다른 sub-skill의 *각 차수 진입 직전* mandatory blocking gate 역할. ## Triggers — INTERNAL ONLY. `vision-foresight-futures-wheel` 마스터 또는 basic-v1·domain-v2·temporal-v3·consequence-linker 등 다른 sub-skill이 차수 산출 직전 *자동* 호출. blocking — 본 gate 통과 못하면 차수 산출 진행 차단. 사용자는 직접 호출하지 않음(disable-model-invocation: true). ## Detailed Methodology — 박사님 강화 명령 3종 통합 구현. 핵심 5 protocol: ① **MDE (Minimum Depth Enforcement)** — depth_target ≥6 강제, exception은 "간단히/빠르게" 명시 시만 ② **PRRG (Per-Ring Research Gate) 6 Gates** — Gate_P1_Pre~Gate_P6_Pre, 차수마다 WebSearch ≥3, primary count별 evidence search, historical analog 강제, backlash analog (4차+), paradigm shift analog (5차+), civilizational analog (6차+) ③ **CoER (Chain-of-Evidence Reasoning)** — 각 impact마다 base fact[R-1]→intermediate inference[R-2]→leap to impact[H
-
idoforgod Bundle Vision Foresight Scenarios Scenario Logics SelectionTLDR — INTERNAL Sub-skill. vision-foresight-scenarios 대표 마스터의 4번째 sub-skill. Glenn & TFG V3.0 19장 Section III Schwartz Step 4 + TFG Preparation scenario space + MITRE 4-quadrant case + Defense Markets 4-dimension verbatim 풀 구현. **Scenario Logics Selector AI Agent** — Schwartz Step 4 verbatim: *"select the scenario logics"*. Mandel-Wilson SRI verbatim: *"The team then develops scenario theories or 'logics,' which are differing views of the way the world might work in the future. Each theory takes into account critical drivers and uncertainties."* MITRE case 양식 (PDF Section III verbatim): *"In the law enforcement case, these axes helped define four scenarios of interest: a. High funding, permissive attitudes toward crime / b. High funding, repressive attitudes toward crime / c. Low funding, permissive attitudes toward crime / d. Low funding, repressive attitudes toward crime"*. PDF Section III recommendation: *"Defining a large number of alternative worlds is often neither necessary nor desirable. A smaller set
-
idoforgod Bundle Vision Foresight Scenarios Internal Consistency CheckTLDR — INTERNAL Sub-skill. vision-foresight-scenarios 대표 마스터의 8번째 sub-skill. Glenn & TFG V3.0 19장 Section II "Good scenarios" 3 criteria verbatim 풀 구현. **Internal Consistency Checker AI Agent** — PDF Section II 핵심 verbatim: *"'Good' scenarios are those that are: 1) Plausible (a rational route from here to there that make causal processes and decisions explicit); 2) Internally consistent (alternative scenarios should address similar issues so that they can be compared); and 3) Sufficiently interesting and exciting to make the future 'real' enough to elicit strategic responses."* 본 sub-skill은 narrative writer의 output을 받아 3 criteria 각각 자동 audit. ① Plausible: causal chain explicit·rational route·decision points 보임 ② Internally consistent: 4-5 scenarios가 *similar issues* address (cross-comparison 가능)·각 scenario 내부 모순 없음 ③ Sufficiently interesting: vivid·surprises·strategic response trigger 가능. Violations 발견 시 Narrative Writer 재호출(최대 2회) 또는 master 에스컬레이션. **DEFAULT: 3 criteria 자동 audit + 5-level Likert per criterio
-
idoforgod Bundle Vision Foresight Environmental Scanning Quest WorkshopTLDR — Burt Nanus·Slaughter (1990) QUEST 4-phase 워크숍 풀 구현 INTERNAL SUB-SKILL. ## Triggers — 사용자 직접 trigger 차단 (disable-model-invocation: true). 대표 스킬 Cycle D에서 AI QUEST Workshop Participants Agent (12~15 페르소나)가 자동 호출. ## Detailed Methodology — Phase 1 Preparation·Phase 2 All-day Scanning Workshop·Phase 3 2-section Report·Phase 4 Half-day Strategic Options. 12~15 페르소나 자동 시뮬레이션. 4 Alternative Scenarios (Plausible·Possible·Wild Card·Preferred). 외부 사람 동원 X.
-
idoforgod Bundle Vision Foresight Scenarios Driving Forces IdentificationTLDR — INTERNAL Sub-skill. vision-foresight-scenarios 대표 마스터의 2번째 sub-skill. Glenn & TFG V3.0 19장 Section III Schwartz Step 2 + TFG Preparation 풀 구현. **Driving Forces Identifier AI Agent** — Peter Schwartz Step 2 verbatim: *"identify the key forces and trends in the environment"* + Schwartz *"driving forces as the building blocks of scenarios"*. TFG: *"Define the scenario space. A scenario study begins by defining the domain of interest. Given a clear statement of the domain, analysts list key driving forces thought to be important to the future of the domain."* (Section III verbatim). MITRE case 양식: *"driving forces of law enforcement funding and social attitudes toward crime were defined as ultimately important. To the degree possible, these driving forces should be independent 'axes' in a scenario space."* Mandel & Wilson SRI verbatim: *"The team then analyzes forces that will shape the future business environment, both from within their own industry (competition) and outside of it (social, political, econ
-
idoforgod Bundle Vision Foresight Environmental Scanning Issues ManagementTLDR — Renfro (1993) Issues Management 4단계 cycle 풀 구현 INTERNAL SUB-SKILL. ## Triggers — 사용자 직접 trigger 차단 (disable-model-invocation: true). 대표 스킬 Step 5에서 AI Issues Committee Agent (10 페르소나)가 자동 호출. ## Detailed Methodology — Renfro 4 stages (Identify·Research·Evaluate·Strategy). PI Chart secret balloting + Top 3-5 Consensus + 4 task forces (Recommendation·Development·Research·Directed Scanning). OPIN 8 questions·23 techniques matrix 자동.
-
idoforgod Bundle Vision Foresight Scenarios Importance Uncertainty RankingTLDR — INTERNAL Sub-skill. vision-foresight-scenarios 대표 마스터의 3번째 sub-skill. Glenn J.C. & TFG (2009) *Futures Research Methodology V3.0* 19장 Section III Schwartz Step 3 풀 구현. **Importance × Uncertainty Ranker AI Agent** — Schwartz P. (1991) *The Art of the Long View* pp. 226-234 Seven Step Process Step 3 verbatim: *"rank the driving forces and trends by importance and uncertainty"*. 핵심 도구 — 2×2 matrix (Importance axis × Uncertainty axis) → 4 quadrant 분류: ① **High Importance + High Uncertainty = Critical Drivers** (scenario axes 후보, Step 4 logics 선택 input) ② **High Importance + Low Uncertainty = Predetermined Elements** (모든 scenario에 공통, Wack 1985 / van der Heijden 1996) ③ **Low Importance + High Uncertainty = Noise** (제외) ④ **Low Importance + Low Uncertainty = Background**. Top-2 Critical Drivers가 4-quadrant scenarios의 axes로 권장 (Schoemaker 1995). **결정론 도구** — `iu_ranker.py` (이 폴더 내) — verbatim·Likert anchor·quadrant 임계값(3.5)·median/mean/std 집계·Top-2 정렬 규칙·입력 schema 검증·ASCII 2×2 matrix 렌더링을 LLM이 자연어로 재추론하지 않도록
-
idoforgod Bundle Vision Foresight Environmental Scanning Weak Signal TemplateTLDR — Gordon·Glenn (2009) 02장의 10·13-field 템플릿 + weak signal 패턴 추출 풀 구현 INTERNAL SUB-SKILL. ## Triggers — 사용자 직접 trigger 차단 (disable-model-invocation: true). 대표 스킬 Step 3에서 AI Analyst Agent가 자동 호출. ## Detailed Methodology — 10-field KOC + 13-field UNDP 템플릿 자동 변환. 9 Gordon-Glenn 도메인 + Spirituality multi-classification. 5 weak signal pattern (frequency spike·cross-domain·신규 actor·consequence shift·status convergence) 자동 detection.
-
andrewnggirl Skill Finance Scenario AdvisorUse when a finance-related Agent Skill does not clearly fit one sub-scenario and needs intake classification, decision-impact assessment, risk boundary review, and recommendations for whether to evaluate it as fundraising, quant trading, stock trading, securities research, banking workflow, financial education, or financial data analysis.
-
andrewnggirl Skill Financial Data Analysis AgentUse when analysts, finance teams, or operators need to clean financial datasets, check data quality, analyze revenue, margin, cash flow, customer value, anomalies, or trends, and produce traceable management-ready insights with confidence caveats.
-
bunsdev Bundle Long Running GoalsMaintain long-running, cross-session goals for an agent (Copilot CLI, Claude Code, Codex, etc.) using a declarative goals.md file. Use when the user wants the agent to remember objectives across sessions, when they say "let's set a goal", "remember to keep working on X", "what are we working on?", "resume the goal", or asks for persistent objectives that survive context resets and compaction.
-
sunjongos Skill Notebooklm MCP Skill---
-
sol1986 Bundle Mindstudio Generate Image Block SkillWrite, configure, and optimize prompts for MindStudio's Generate Image block. Use this skill whenever someone wants to generate an image inside a MindStudio workflow, write an image generation prompt, use a variable in an image prompt, choose an AI image model, access the output image URL, resize or transform an image via CDN, or connect image generation to a downstream block. Always use this skill for any MindStudio image generation task, even if the request seems simple.
Frequently asked questions
What are AI & ML agent skills?
AI & ML agent skills cover the machine-learning workflow itself: writing and evaluating prompts, building RAG pipelines, running evals, and wiring up model APIs. Each one is a SKILL.md file your agent loads on demand, so the know-how travels across Claude Code, Cursor, and 60+ agents.
Which AI & ML skills are most installed?
Popular AI & ML skills on SkillMD right now include statsmodels, scikit-learn, vision-foresight-futures-wheel-basic-v1. Rankings shift as installs change; sort this page by "Most installs" for the live list.
Do AI & ML skills work with Claude Code and Cursor?
Yes. Every skill here ships as a SKILL.md file, an open format that works in Claude Code, Claude.ai, Cursor, Codex, Windsurf, and 60+ other agents. Install one with npx skillmds@latest add <owner>/<name>, or copy the file into your agent's skills directory.