Board Meeting
Multi-agent board meeting protocol for strategic decisions. Runs a structured 6-phase deliberation: context loading, independent C-suite contributions
What Is Board Meeting?
The Board Meeting skill is a multi-agent deliberation protocol designed for large language models, providing a structured framework for simulating high-level executive board meetings. Tailored for strategic business decisions, it orchestrates contributions from simulated C-suite executives—such as CTO, CFO, COO—in a controlled, phase-based environment to ensure diversity of perspective and prevent groupthink. The protocol is structured into six rigorous phases, from initial context loading to final decision extraction, ensuring that each executive’s input is independent and that deliberations remain focused, transparent, and actionable. Board Meeting is particularly effective for users seeking thorough, multi-perspective executive analysis on complex strategic questions.
Why Use Board Meeting?
Traditional group decision-making processes are susceptible to biases, groupthink, and the loss of minority perspectives. The Board Meeting skill addresses these issues by enforcing independent contributions from each simulated executive before any synthesis occurs. This methodology ensures that all viewpoints—especially dissenting or minority opinions—are captured and considered. The structured protocol is valuable for:
- Strategic planning: Simulate a boardroom deliberation on expansion, mergers, product launches, or risk management.
- Design sprints: Gather multi-disciplinary inputs on a design or technical challenge.
- Scenario analysis: Quickly generate, critique, and synthesize executive-level responses to hypothetical situations.
- Governance: Documented, repeatable decision-making with clear rationale and traceability.
By leveraging Board Meeting, organizations and individuals can access a process-driven, AI-powered “virtual board” to support critical business decisions.
How to Get Started
To invoke the Board Meeting skill, use the following command syntax:
/cs:board [topic]For example:
/cs:board Should we expand to Spain in Q3?Behind the scenes, the protocol proceeds through six phases:
- Context Gathering: Loads organizational context and previous decision summaries, resets session state, and presents the agenda and relevant C-suite roles for founder (user) confirmation.
- Independent C-suite Contributions: Each activated executive role independently drafts their perspective—no cross-pollination or shared memory—to prevent bias.
- Critic Analysis: A designated critic agent reviews all executive contributions, highlighting strengths, weaknesses, and identifying conflicts or missing viewpoints.
- Synthesis: The protocol synthesizes inputs into a unified, actionable summary, preserving unique perspectives and surfacing disagreements.
- Founder Review: The user (acting as founder or board chair) reviews the synthesized options, requests clarifications, and can prompt further exploration.
- Decision Extraction: The protocol extracts clear decisions and next steps, documenting rationale and archiving outcomes for future reference.
Example:
Minimal Invocation
/cs:board Should we pivot our product strategy to target enterprise customers?The Board Meeting skill will present the agenda, activate relevant C-suite agents (e.g., CEO, CPO, CRO), and commence the 6-phase protocol.
Key Features
- Multi-agent architecture: Simulates realistic executive deliberations with separate memory and reasoning for each C-suite role.
- 6-phase structured protocol: Ensures comprehensive, bias-resistant decision-making.
- Role selection logic: Chief of Staff dynamically activates only the most relevant executive roles for each topic.
- Two-layer memory system: Draws from curated company context and past board decisions—never raw transcripts—to inform current deliberations.
- Actionable outputs: Produces concise decision summaries, rationale, and next steps suitable for direct implementation or documentation.
- Extensible and open source: MIT-licensed and modular for easy integration or adaptation into custom workflows.
Best Practices
- Curate company context: Regularly update
memory/company-context.mdandmemory/board-meetings/decisions.mdto keep deliberations relevant and accurate. - Select roles judiciously: Activate only those executive roles directly relevant to the topic to maintain focus and avoid information overload.
- Encourage dissent: Use the critic analysis phase to surface and address disagreements or minority viewpoints.
- Document decisions: Archive all outputs for future reference and traceability, supporting organizational learning and governance.
- Iterate when needed: The founder review phase can be used to request further analysis, additional roles, or clarification before finalizing decisions.
Important Notes
- Strict isolation: During the independent contribution phase, executives do not access each other’s drafts to ensure unbiased input.
- No raw transcript loading: Only curated summaries from prior meetings are loaded—never full transcripts—to maintain context relevance and protect privacy.
- Session state reset: Each invocation resets state, preventing bleed from previous conversations and ensuring clean deliberation.
- Role activation logic: Not all C-suite roles are activated for every topic; the Chief of Staff selects based on relevance.
- Open source: The Board Meeting skill is available under the MIT license at GitHub for inspection, extension, and contribution.
By following these guidelines and understanding the protocol’s structure, users can leverage Board Meeting to simulate robust, diverse, and transparent executive decision-making within large language model environments.
More Skills You Might Like
Explore similar skills to enhance your workflow
DDD
Domain-driven development skills that also include Clean Architecture, SOLID principles, and design patterns
Phase 1: Identify Balance Domain
If no argument, ask the user which system to check
Flux Image
Flux Image automation for generating and processing high-quality creative visual content
Core Principle
- ~100 customers (repeat customers = product-market fit)
SQL Queries
Generate SQL queries from natural language descriptions. Supports BigQuery, PostgreSQL, MySQL, and other dialects. Reads database schemas from
LangChain & LangGraph Architecture
Master modern LangChain 1.x and LangGraph for building sophisticated LLM applications with agents, state management, memory, and tool integration