Round 75: standardize 05-domain prompt contracts + eval anti-drift

Six-lens panel consensus (DR–DT): add Purpose/Scope/Acceptance/Falsifiability
preambles to all seven 05-domain prompts with thinking/workflow cross-links;
add test_domain_prompts_eval_harness.py; extend test_prompt_cross_links.py;
sync governance docs to Round 75.

Author: johnteee

README.md

@@ -21,7 +21,7 @@ Full library docs: [reflective-prompt-library/README.md](reflective-prompt-libra
 ## Governance
 
 - **Contributing:** [CONTRIBUTING.md](CONTRIBUTING.md) — quality gates, routing maintenance (R8–R12), `make all`
-- **Panel record:** [multi-agent-panel-consensus](reflective-prompt-library/plans/multi-agent-panel-consensus-2026-06-25.md) — six-lens Socratic consensus (Rounds 1–74)
+- **Panel record:** [multi-agent-panel-consensus](reflective-prompt-library/plans/multi-agent-panel-consensus-2026-06-25.md) — six-lens Socratic consensus (Rounds 1–75)
 - **Operator playbook:** [GLOSSARY.md](reflective-prompt-library/GLOSSARY.md) — Governance Maintenance Playbook
 
 The repository contains:

reflective-prompt-library/05-domain/business-strategy.md

@@ -2,6 +2,24 @@
 
 Use this for business models, product positioning, transition strategy, or venture review.
 
+## Purpose
+
+Review business models, product positioning, transition strategy, and venture proposals. Primary workflow surfaces: `reflective-brief` and `reflective-research`. Pairs with `01-thinking/counterargument.md`, `01-thinking/falsifiability.md`, and `01-thinking/socratic-reviewer.md`.
+
+## Scope
+
+- In scope: value proposition, alternatives, moat, kill criteria, counterarguments.
+- Out of scope: implementation delivery (`reflective-implement`) without approved spec.
+
+## Acceptance Criteria
+
+- Paid demand and distribution costs challenged explicitly.
+- Kill criteria and MVP path stated before scaling recommendations.
+
+## Falsifiability
+
+Name one assumption that, if false, would invalidate the proposed business model.
+
 ```markdown
 你是 Strategy Reviewer。請分析以下商業 / 產品方案。
 

reflective-prompt-library/05-domain/creative-template.md

@@ -2,6 +2,28 @@
 
 Use this for advertising creative, templates, animation descriptions, and reproducible creative specs.
 
+## Purpose
+
+Convert natural-language creative intent into verifiable, replayable creative specs. Primary workflow surface: `reflective-spec-plan`. Pairs with `01-thinking/falsifiability.md` and `01-thinking/why-what-how-done.md`.
+
+## Scope
+
+- In scope: scene breakdown, constraints, validation rules, brand-safety checks, fallback paths.
+- Out of scope: unvalidated final renders (`reflective-implement`) without preview gates.
+
+## Acceptance Criteria
+
+- Animation and visual constraints expressed as testable fields (duration, easing, fallback).
+- Brand-safety and failure cases listed before handoff to execution.
+
+## Falsifiability
+
+State what preview or validation result would reject the creative spec.
+
+## Human Review
+
+Require human approval before publishing creative output that affects brand or external audiences.
+
 ```markdown
 你是 Creative Template Architect。請把自然語言創意轉成可驗證、可重播、可約束的 creative spec。
 

reflective-prompt-library/05-domain/deep-analysis.md

@@ -2,6 +2,24 @@
 
 Use this for deeper conceptual, architectural, or critical-thinking analysis.
 
+## Purpose
+
+Deep conceptual, architectural, and critical-thinking analysis beyond quick answers. Primary workflow surfaces: `reflective-research` and `reflective-spec-plan`. Pairs with `01-thinking/critical-thinking-check.md`, `01-thinking/counterargument.md`, and `01-thinking/socratic-reviewer.md`.
+
+## Scope
+
+- In scope: problem reframing, assumptions, counterarguments, minimal vs full options.
+- Out of scope: unreviewed delivery (`reflective-implement`) on high-blast-radius conclusions.
+
+## Acceptance Criteria
+
+- Counterarguments and weak points surfaced before recommendation.
+- Acceptance criteria stated for both minimal and full solution paths.
+
+## Falsifiability
+
+Name one counterargument that would flip the recommended decision.
+
 ```markdown
 請以架構師 + 批判性思考教練的角度分析以下問題。
 

reflective-prompt-library/05-domain/high-risk.md

@@ -2,6 +2,28 @@
 
 Use this for security, privacy, money, destructive operations, and production-sensitive work.
 
+## Purpose
+
+Gate security, privacy, billing, destructive, and production-sensitive work before irreversible action. Primary workflow surfaces: `reflective-risk` and `reflective-review`. Pairs with `01-thinking/critical-thinking-check.md`, `01-thinking/counterargument.md`, and `01-thinking/falsifiability.md`.
+
+## Scope
+
+- In scope: threat model, assumption audit, rollback/dry-run plan, go/no-go framing.
+- Out of scope: silent execution of irreversible changes (`reflective-implement` without review).
+
+## Acceptance Criteria
+
+- Human confirmation required before irreversible or high-blast-radius recommendations.
+- Rollback and dry-run paths explicit when delivery is contemplated.
+
+## Falsifiability
+
+Name one threat or failure mode that would change the go/no-go decision.
+
+## Human Review
+
+Require human approval before any irreversible recommendation or production-facing change.
+
 ```markdown
 這是高風險任務。請啟用 High-risk Review Mode。
 

reflective-prompt-library/05-domain/learning-coach.md

@@ -2,6 +2,24 @@
 
 Use this for learning languages, engineering, philosophy, science, or any skill that needs practice and validation.
 
+## Purpose
+
+Turn learning goals into practiceable, assessable, reflective study plans. Primary workflow surface: `reflective-brief`. Pairs with `01-thinking/falsifiability.md` and `01-thinking/why-what-how-done.md`.
+
+## Scope
+
+- In scope: skill decomposition, mastery gates, active recall, spaced repetition, remediation.
+- Out of scope: repository implementation (`reflective-implement`) or formal spec tickets (`reflective-spec-plan`).
+
+## Acceptance Criteria
+
+- Each practice task has an observable mastery gate before advancing.
+- Failure signals and misconception remediation paths are explicit.
+
+## Falsifiability
+
+State what performance evidence would show the learner is not ready for the next stage.
+
 ```markdown
 你是 Reflective Learning Coach。請幫我把以下學習目標轉成可練習、可驗收、可反思的計畫。
 

reflective-prompt-library/05-domain/research.md

@@ -2,6 +2,24 @@
 
 Use this for research reports, trend analysis, platform comparison, and evidence mapping.
 
+## Purpose
+
+Structure research reports, trend analysis, platform comparison, and evidence mapping. Primary workflow surface: `reflective-research`. Pairs with `01-thinking/falsifiability.md`, `01-thinking/critical-thinking-check.md`, and `01-thinking/counterargument.md`.
+
+## Scope
+
+- In scope: research question, evidence map, competing views, blind spots, decision framework.
+- Out of scope: repository edits (`reflective-implement`) or ticket slicing (`reflective-spec-plan`).
+
+## Acceptance Criteria
+
+- Claims separated into supported, inferred, unverified, and needs fresh verification.
+- Blind-spot output names at least one gap no competing view touched.
+
+## Falsifiability
+
+State what new evidence would overturn the strongest claim in the synthesis.
+
 ```markdown
 你是 Research Analyst。請對以下主題做結構化研究分析。
 

reflective-prompt-library/05-domain/writing-article.md

@@ -2,6 +2,24 @@
 
 Use this to write essays, articles, reports, or argument-driven long-form content.
 
+## Purpose
+
+Draft argument-driven essays, articles, and long-form reports with explicit thesis structure. Primary workflow surfaces: `reflective-brief` and `reflective-review`. Pairs with `01-thinking/counterargument.md`, `01-thinking/critical-thinking-check.md`, and `01-thinking/socratic-reviewer.md`.
+
+## Scope
+
+- In scope: thesis, claims, evidence needs, counterarguments, revision checklist.
+- Out of scope: publishing or repo changes without review (`reflective-review` gate).
+
+## Acceptance Criteria
+
+- Central thesis and per-paragraph function stated before draft prose.
+- Counterarguments addressed for each major claim.
+
+## Falsifiability
+
+Name one counterargument that would require rewriting the central thesis.
+
 ```markdown
 你是嚴格的 Editor + Argument Architect。請幫我把以下主題寫成有論證結構的文章。
 

reflective-prompt-library/PROJECT_KNOWLEDGE.md

@@ -75,6 +75,7 @@ deferred promotions are recurrence-gated — see [panel backlog](plans/multi-age
 > Pointers to the causal trail — plans, reflections, tests, commits. Detail is
 > not duplicated here; this is a map, not an archive.
 
+- 2026-06-25 Round 75 panel — standardize `05-domain/` prompt contracts (Purpose/Scope/Acceptance/Falsifiability) + thinking/workflow cross-links + `test_domain_prompts_eval_harness.py` → [record](plans/multi-agent-panel-consensus-2026-06-25.md)
 - 2026-06-25 Round 74 panel — standardize `03-context/` prompt contracts (Purpose/Scope/Acceptance/Falsifiability) + thinking/workflow cross-links + `test_context_prompts_eval_harness.py` → [record](plans/multi-agent-panel-consensus-2026-06-25.md)
 - 2026-06-25 Round 73 panel — standardize `04-agent/` prompt contracts (Purpose/Scope/Acceptance/Falsifiability) + thinking/workflow cross-links + `test_agent_prompts_eval_harness.py` → [record](plans/multi-agent-panel-consensus-2026-06-25.md)
 - 2026-06-25 Round 72 panel — standardize `00-core/` prompt contracts (Purpose/Scope/Acceptance/Falsifiability) + eval_harness anti-drift → [record](plans/multi-agent-panel-consensus-2026-06-25.md)

reflective-prompt-library/README.md

@@ -30,7 +30,7 @@ Pick **Strictness L1–L6** first (`skills/reflective-dispatch/SKILL.md`, [GLOSS
 
 ## Governance Panel Record
 
-Multi-agent Socratic consensus on project goals and the nine skills (Rounds 1–74, options A–DQ) is recorded in [plans/multi-agent-panel-consensus-2026-06-25.md](plans/multi-agent-panel-consensus-2026-06-25.md). Run `make all` before claiming routing or governance changes are verified.
+Multi-agent Socratic consensus on project goals and the nine skills (Rounds 1–75, options A–DT) is recorded in [plans/multi-agent-panel-consensus-2026-06-25.md](plans/multi-agent-panel-consensus-2026-06-25.md). Run `make all` before claiming routing or governance changes are verified.
 
 ## Directory Map