Advisory Committee

What This Skill Does

The Advisory Committee skill is a strategic orchestration tool designed to act as a 4-agent board of directors within your OpenClaw environment. Rather than relying on a single AI perspective, this skill mimics a professional corporate governance process by running four specialized agents—CS, Marketing, Finance, and Tech—in sequence to evaluate any business or project proposal. Each agent brings a distinct bias: the User Champion focuses on customer delight, the Growth Hacker looks at acquisition potential, the CFO audits financial viability, and the CTO assesses technical debt and feasibility. The skill aggregates these independent votes into a final recommendation, ensuring a holistic decision-making process for complex dilemmas.

Installation

To integrate this skill into your OpenClaw environment, execute the following command in your terminal: clawhub install openclaw/skills/skills/jini92/advisory-committee Ensure you have the latest version of the OpenClaw agent runtime installed to support sequential agent execution and log table updates.

Use Cases

This skill is best utilized when you are facing high-stakes decisions that involve cross-functional trade-offs. Examples include:

• Determining if a new product feature aligns with both customer demand and technical capacity.
• Evaluating the potential ROI and market impact of a proposed marketing campaign.
• Deciding whether to pivot an existing project or continue iterating on current infrastructure.
• Prioritizing long-term growth initiatives versus immediate operational fixes.

Example Prompts

1. "Advisory Committee, should we pivot our entire infrastructure to a decentralized database architecture for Q3?"
2. "Please evaluate this proposal: I want to spend $5,000 on a viral influencer marketing campaign for our new open-source tool. Is this a good use of resources?"
3. "Run the committee on this idea: Adding an AI-powered chatbot to our support page to handle tier-1 inquiries."

Tips & Limitations

To get the most out of the committee, provide highly specific context in your proposal. If you are vague about budget, current tech debt, or specific user segments, the agents will have to make assumptions, which may lead to generic feedback. Note that this skill consumes significantly more tokens than a single-agent task, as it runs four independent inference cycles. It is recommended to save the output in your project documentation for later reference, as the committee acts as an excellent audit trail for strategic shifts.