decision-bias-quality-control
Decision Bias Quality Control
Overview
使用此技能來執行「決策品管」,把重大決策從直覺推進到可追溯、可檢查、可比較的判斷流程。
遵循兩個核心原則:
- 維持「雙軌輸出」:每次都同時輸出質性洞察與量化評分。
- 檢查「提案過程」而非只看「提案結論」:用 12 問檢查清單找出偏誤來源。
先閱讀下列參考檔:
- 核心原則與來源:references/01-source-principles.md
- 12 問題庫:references/02-12-question-bank.md
- 三種模式流程:references/03-mode-workflows.md
- 評分與門檻:references/04-scoring-thresholds.md
- 統一輸出模板:references/05-output-templates.md
Input Contract
接收下列結構化輸入;欄位不足時走 Data Sufficiency Gate。
mode:proposal-review | meeting-facilitation | personal-coaching | autodecision_statement:string(必要)decision_context:string(必要)proposal_snapshot:string(提案審查時建議提供)decision_size_factors:financial_impact:1|2|3reversibility:1|2|3org_blast_radius:1|2|3
Mode Routing
若 mode=auto,依需求語意路由:
- 出現「審查提案、投資案、核准、go/no-go」等語意時,路由
proposal-review。 - 出現「主持、引導討論、會議流程、追問順序」等語意時,路由
meeting-facilitation。 - 出現「我該怎麼選、個人抉擇、職涯選擇」等語意時,路由
personal-coaching。
若語意混合,先以 proposal-review 執行主流程,並在附錄提供會議引導問題。
Data Sufficiency Gate
先檢查必要欄位:
decision_statementdecision_context
再檢查評分所需欄位:
decision_size_factors.financial_impactdecision_size_factors.reversibilitydecision_size_factors.org_blast_radius
缺資料時,輸出 MissingDataOutput:
missing_fieldswhy_neededquestions_to_usertemporary_assumptionrisk_of_assumption
預設規則:
- 若三因子缺漏且使用者仍要求繼續,暫定
decision_size=Medium,並標註風險。
Output Contract (Mandatory Dual Track)
每次輸出都必須同時包含以下兩軌,不得省略其中任一軌。
A. 質性軌
bias_diagnosis: 偏誤診斷摘要(依 12 問分組)key_risks: 3-5 個核心風險(附證據來源)recommended_actions: 3-5 個可執行行動(含優先級)
B. 量化軌
question_scores: 12 題逐題分數(0|1|2)與簡要理由group_scores: 三組標準化分數(0-100)weighted_total_score: 加權總分(0-100)decision_size:Low|Medium|Highrisk_level:Green|Yellow|Redthreshold_profile_used: 對應門檻表
Scoring Rules
依 references/04-scoring-thresholds.md 計算,使用固定模型:
- 題目分數:每題
0|1|2 - 分組加權:
- Q1-3 =
30% - Q4-9 =
40% - Q10-12 =
30%
- Q1-3 =
- 分組先標準化至 0-100,再算加權總分
- 動態門檻依決策大小調整風險等級
Execution Workflow
- 確認模式與決策邊界。
- 進行 Data Sufficiency Gate。
- 逐題執行 12 問,收集證據與評分。
- 計算分組分數、總分、風險等級。
- 輸出雙軌結果與 mode-specific 結論。
- 附上「下一步驗證行動」與「可逆/不可逆提醒」。
Mode-Specific Decision Output
proposal-review: 輸出go | conditional-go | no-go。meeting-facilitation: 輸出會議腳本、追問順序、決策收斂規則。personal-coaching: 輸出選項比較、偏誤提醒、48 小時驗證行動。
詳見 references/03-mode-workflows.md。
Quality Rules
- 不可跳過 12 問中的任何一題。
- 不可輸出無證據支持的結論。
- 不可只給結論不給計算過程。
- 不可把單一觀點誤當團隊共識。
- 必須標註假設與不確定性來源。
Suggested Prompt
Use $decision-bias-quality-control to evaluate this high-stakes decision in dual-track mode with full 12-question scoring and dynamic risk thresholds.
More from timlai666/skills
investment-research-prompts
Use when the user needs stock screening, portfolio risk review, dividend portfolio design, pre-earnings analysis, industry competition analysis, DCF valuation, technical analysis, or stock trend/anomaly detection. Trigger on requests like 選股, 投資組合風險, 股息策略, 財報前瞻, DCF 估值, 技術面分析, 產業競爭研究, 趨勢識別, or investment research prompts.
22landing-page-studio
Use when creating high-conversion landing pages with modern visual effects, including SVG/WebGL animation, autonomous multi-iteration optimization, and dual output modes (single-file HTML or React project). Trigger on requests like landing page, LP, hero section build, animated marketing page, conversion page redesign, or style variants A-L/custom.
17bcg-growth-share-matrix
Use when analyzing a group, business-unit, product-line, or brand portfolio with the BCG growth-share matrix, especially when the task involves relative market share, market growth, quadrant classification, capital allocation, portfolio prioritization, or deciding whether to invest, maintain, harvest, reposition, or exit.
17data-analysis-workflow
Use when planning or executing a full data analysis workflow, including schema inspection, data quality audit, data cleaning, EDA, relationship analysis, feature engineering, modeling, evaluation, and report generation. Trigger on requests like 資料分析流程, EDA 到建模, 數據分析規劃, 分析報告產出, or end-to-end analytics workflow.
16psychological-trigger-marketing
Use when generating high-conversion marketing angles, campaign hooks, landing page messaging, promotional copy directions, social post hooks, or CTA concepts with psychological triggers such as FOMO, justification, desire, priming, anchoring, and framing, especially for offers, launches, limited-time promotions, and Traditional Chinese marketing for Taiwan audiences.
16service-innovation-workshop
Use when turning a service innovation problem into concept options, prototype testing, and risk checks, especially for 服務創新, 創新機會, 價值共創, 服務原型, SCAMPER, 購買者效用矩陣, 創新流程, 創新困境, or 服務體驗工程. Trigger when the task asks for practical innovation directions, concept generation, opportunity framing, or validation planning rather than only concept definitions.
16