executive-protocol
Enterprise executive decision frameworks, memo templates, board protocol, RACI, OKR, escalation rules, dissenting-opinion format.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
메뉴
Enterprise executive decision frameworks, memo templates, board protocol, RACI, OKR, escalation rules, dissenting-opinion format.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
AI governance — EU AI Act Article 9, NIST AI RMF, ISO/IEC 42001 — risk classification, lifecycle gates, model cards, HITL policy, AI incident response.
Crisis response toolkit — classification, trigger thresholds, escalation tiers, war-room activation, rapid liquidity stress, supply-chain alternate-source, holding-statement, AAR.
Adversarial red-team debate protocol — when to use it, the 4-step structure, role-pair recipes, Referee scoring, termination conditions, MAS failure mitigations.
Enterprise risk management — COSO ERM 2017 + ISO 31000 — risk taxonomy, appetite statement, top-risk register, KRIs, Bow-Tie, three-lines-of-defense.
Deterministic financial tools — WACC, NPV, IRR, payback, EVA, real-options, Monte Carlo, sensitivity, covenant/liquidity checks. The Financial Hardcoding Directive.
M&A playbook — deal thesis, screening, financial/commercial/legal diligence, valuation methods, real-options on deal structure, integration playbook, post-deal KPIs.
| name | executive-protocol |
| description | Enterprise executive decision frameworks, memo templates, board protocol, RACI, OKR, escalation rules, dissenting-opinion format. |
| allowed-tools | ["Read","Write","Glob","Grep"] |
The shared operating protocol for all C-suite agents in .claude/agents/. Every output from a single-domain executive or an orchestrator MUST conform to one of the templates below.
# [Topic] — [Role] Analysis — YYYY-MM-DD
## Executive Summary
[2–3 sentence recommendation with confidence level: High / Medium / Low and the reason]
## Situation Assessment
[Current state; what's at stake; what we know / don't know]
## Options Considered
### Option A — [Name]
- Description
- Pros (with evidence)
- Cons (with evidence)
- Score: X/10 on [decision framework]
### Option B — [Name]
[same structure]
### Option C — [Name] (always include "do nothing / status quo" or "kill / return capital")
[same structure]
## Recommendation
[Clear, actionable recommendation with rationale and trade-off statement]
## Decision Framework Scoring
| Criterion | Weight | Option A | Option B | Option C |
|---|---|---|---|---|
## Next Steps
| # | Action | Owner | Deadline | Success Criterion |
|---|---|---|---|---|
## Risk Factors & Mitigations
| Risk | Probability | Impact | Mitigation | Trigger |
|---|---|---|---|---|
## Assumptions That Could Be Wrong
[List the 3–5 assumptions whose failure would change the recommendation]
## HITL / Approvals Required
[Who must approve before execution; what decisions remain with the human]
---
Filed by: [role-slug] | Date: YYYY-MM-DD
Saved to: output/[domain]/[topic]-YYYY-MM-DD.md
Used by boardroom, mna-cockpit, crisis-warroom, capital-allocation.
# Board Meeting — [Topic] — YYYY-MM-DD
## Agenda
[Single-sentence decision frame]
## 1. Situation Brief
[Context, scope, known facts, known unknowns]
## 2. Functional Perspectives
[For each attending executive — 3 to 5 perspectives — analyze through their decision framework]
### [Role] Perspective
- Their framework's top scoring criteria
- Recommendation
- Top risks they see
- What evidence would change their position
## 3. Points of Agreement
[Be specific; not "everyone wants to win"]
## 4. Points of Tension
| Tension | Side A | Side B | Resolution Path |
|---|---|---|---|
(If unanimous, explicitly note "no material tension surfaced — flag for groupthink review")
## 5. Board Recommendation
[Synthesized recommendation with confidence level + the reason]
## 6. Action Items
| Action | Owner | Deadline | Success Criterion |
## 7. Open Questions / HITL
[What requires human decision before execution]
## 8. Dissenting Opinions
[Per the Dissenting Opinion Format below — recorded verbatim, never paraphrased away]
---
Filed by: [orchestrator-slug] | Date: YYYY-MM-DD
| Decision | R (Responsible) | A (Accountable) | C (Consulted) | I (Informed) |
|---|---|---|---|---|
| Pricing change (material) | cro | cfo | cmo, cpo | ceo |
| New market entry | cso | ceo | cfo, cro, cmo, clo | board |
| M&A signing | mna-cockpit / clo | ceo | cfo, cso, coo, board | all |
| Capital project > threshold | capital-allocation | ceo | all relevant | board |
| AI deployment (high-risk class) | caio | ceo | clo, chief-compliance-officer, ciso | board |
| Cyber-incident response (SEV-1) | ciso / crisis-warroom | ceo | clo, chief-communications-officer, cfo | board |
| Layoff / RIF | chro | ceo | clo, cfo, chief-communications-officer | board |
| Regulatory filing (material) | chief-compliance-officer / clo | ceo | cfo | board |
| Brand reposition | cmo | ceo | chief-communications-officer, cpo | all |
| Data-platform rebuild | cdo | cto | cio, ciso, cfo | exec team |
| Climate transition plan | chief-sustainability-officer | ceo | cfo, csco, cto | board |
| Executive comp (NEO) | chro | board comp committee | ceo, cfo | board |
## Q[N] 20XX OKRs — [Function / Firm]
### Objective 1: [Ambitious qualitative statement]
- KR1: [Metric] from [baseline] to [target] by [date]
- KR2: [Metric] from [baseline] to [target] by [date]
- KR3: [Metric] from [baseline] to [target] by [date]
Progress: [0.0–1.0; stretch = 0.7] | Owner: [exec-slug] | Risk: [color]
Cadence: monthly mid-cycle check, quarter-end scoring, retrospective.
| Scope | Authority | Process |
|---|---|---|
| Within single domain, within budget, reversible | Domain exec decides | Decision log |
| Cross-domain or above $/% threshold | boardroom (Quick Consult or Full) | Memo |
| Material capital or M&A | capital-allocation or mna-cockpit | Full debate protocol |
| Material legal / regulatory exposure | clo go/no-go authority | Legal memo + CEO notification |
| Risk appetite breach | chief-risk-officer escalation | Board notification |
| Crisis triggers | crisis-warroom activation | CEO + board notification |
| Board-reserved matters | Board | Per charter (committee structure) |
| Level | Criteria |
|---|---|
| High | Multiple independent evidence sources; tested logic; no critical-assumption risk; senior team aligned; analogous precedent |
| Medium | Some evidence; some critical assumptions still unverified; mixed analogous precedent; surfaceable disagreement |
| Low | Mostly hypothesis; key data missing; novel situation; team disagreement; high reversibility cost |
Stating confidence is mandatory. "Not confident" is a legitimate output; "false confidence" is not.
When a perspective disagrees with the synthesized recommendation:
### Dissent — [role-slug]
- Position: [What I think the right action is, in one sentence]
- Reasoning: [3–5 bullets — evidence, framework, precedent]
- Specific risk if majority recommendation is taken: [What I think breaks]
- Conditions under which I would change my position: [Evidence or events]
Dissent is preserved verbatim in the meeting minutes. Never paraphrased away. The research doc explicitly recommends preserving dissenting opinions to mitigate ensemble-LLM bias and groupthink.
# Strategy Canvas — [Topic]
## Current Position
| Factor | Our Position (1–10) | Market Average (1–10) | Best-in-Class (1–10) |
|---|---|---|---|
## Target Position (3 years)
[Where we want to move each factor and why]
## Strategic Moves (Four Actions Framework)
- **Raise** — factors to increase above market
- **Create** — new factors competitors don't offer
- **Reduce** — factors to decrease below market
- **Eliminate** — factors to stop investing in
## Trade-offs Made Explicit
[What we are choosing not to do, and why]
All artifacts go to output/<domain>/<topic-kebab>-YYYY-MM-DD.md per CLAUDE.md.