# COO Operations Prover MCP for AI Agents AI Agent Connect > COO Operations Prover validates infrastructure plans against capacity modeling, failure isolation, cost leverage, process discipline, and accountability. It stops 'best effort' promises by forcing your AI client to prove operational readiness with hard data points like arrival rates, blast radii, and error budgets. ## Overview - **Category:** infrastructure - **Price:** Free - **Endpoint:** https://edge.vinkius.com/vk_preview_k2p6ROIYAs0oOHDMZETbyPYeuNhOK8Zt670VqgWO/ai-agent-connect - **Tags:** coo, operations, capacity, sla, error-budget, scalability, reliability ## Description COO Operations Prover validates infrastructure plans against capacity modeling, failure isolation, cost leverage, process discipline, and accountability. Most operational plans built by AI are full of fluff. They promise scalability without mentioning arrival rates and claim economies of scale without showing a single cost data point. This Connector changes that by acting as a rigorous gatekeeper. It forces your agent to think like a wartime COO who needs to know exactly how a system will behave under pressure. Instead of accepting we will scale as needed, it demands to see the queuing theory math. It won't let a plan pass unless it defines specific bulkheads to prevent contagion and sets concrete error budgets with automated penalties. By using this Connector through the Vinkius catalog, you move from aspirational best effort plans to execution-ready blueprints. It catches the stuff that usually breaks production, like the case-by-case exceptions that destroy leverage and the lack of clear blast radius limits. You get a verdict matrix that tells you exactly where your plan is thin, whether it is capacity blindness or cost delusion. ## Tools ### validate_coo_operations Validates an operational plan against five axes including capacity, failure isolation, cost, process, and accountability. It returns a verdict matrix highlighting specific operational gaps. ## Prompt Examples **Prompt:** ``` Plan our move to a multi-region setup for 50k concurrent users. ``` **Response:** ``` ### Capacity Blindness Detected Your current plan lacks the required modeling. To be **OPERATIONS_PROVEN**, you must provide: | Metric | Required Data | |---|---| | **Arrival Rate** | $\lambda$ req/s | | **Service Rate** | $\mu$ req/s | | **Utilization** | $\rho$ (Target < 70%) | | **Drain Time** | p99 after burst | Please update the plan with these numbers to proceed. ``` **Prompt:** ``` How much will it cost to scale to 1 million users? ``` **Response:** ``` ### Cost Delusion You mentioned economies of scale but provided no data. Please provide the per-unit cost at the following scale points: * **10k Users:** $\$\$0.XX * **100k Users:** $\$\$0.XX * **1M Users:** $\$\$0.XX Include the mechanism (e.g., committed use discounts, cache hit ratios) driving the decrease. ``` **Prompt:** ``` Write an SLA for our new payment gateway. ``` **Response:** ``` ### Accountability Theater 'Best effort' is not an SLA. To pass validation, your plan must include: * **SLA Target:** 99.9% uptime * **Error Budget:** 43.8 minutes per month * **Automatic Penalty:** Feature deploy freeze if the budget burns Please define these specific consequences to move to **OPERATIONS_PROVEN** status. ``` ## Capabilities ### Model capacity using queuing theory Your agent calculates arrival rates, service rates, and queue behavior to prove the system can handle peak load. ### Define failure isolation limits It forces the identification of bulkheads and blast radius limits to prevent a single failure from cascading. ### Verify cost leverage at scale The tool demands proof of decreasing per-unit costs at three specific scale points to validate economic claims. ### Quantify process discipline It requires SOP coverage percentages and keeps exception rates below a 5% threshold to ensure operational leverage. ### Enforce accountability with penalties Your agent must define an error budget and an automated penalty, such as a feature freeze, instead of best effort promises. ## Use Cases ### Scaling a high-traffic API An engineer asks the agent to plan for 10k req/s. The tool catches capacity blindness by demanding a service rate and queue drain time. ### Reviewing a new cloud migration A manager wants to save costs. The tool forces the agent to show cost leverage at 10k, 100k, and 1M users. ### Designing a multi-tenant database The agent proposes a shared pool. The tool flags contagion risk and demands tenant isolation and blast radius limits. ### Auditing an existing SOP A lead wants to check if case-by-case handling is hurting the team. The tool quantifies the exception rate and demands runbooks. ## Benefits - Stop Best Effort SLAs: Use validate_coo_operations to force your agent to define real error budgets and automated penalties. - Prove Scalability: Move past auto-scaling claims by requiring arrival rates and service rates for every plan. - Limit Blast Radii: Ensure your agent identifies specific bulkheads to prevent one shard failure from taking down the whole system. - Verify Cost Leverage: Get proof of decreasing per-unit costs at three specific scale points instead of just economies of scale. - Enforce Process Discipline: Force your agent to identify SOP coverage and keep exception rates below a 5% threshold. - Identify Fatal Gaps: Get a verdict matrix that names the exact operational axis where your plan is currently failing. ## How It Works The bottom line is you get a battle-tested operational blueprint instead of a list of aspirations. 1. Input your operational plan or system design into your AI client. 2. Call the validation tool to run the plan through the five operational axes. 3. Receive a verdict matrix identifying specific fatal gaps like capacity blindness or contagion risk. ## Frequently Asked Questions **What does the COO Operations Prover do for my infrastructure?** It validates your plans against five operational axes to ensure they won't fail under load. It forces your agent to move past vague promises and provide hard data on capacity, costs, and reliability. **How does this help with scaling?** It forces the AI to model actual arrival rates and service rates instead of just saying we will scale. This ensures your infrastructure can actually handle the traffic you expect. **Can it help with cost management?** Yes, it requires proof of cost leverage at three different scale points to avoid cost delusion. It makes sure your plan actually shows decreasing per-unit costs as you grow. **What is a Contagion Risk in this context?** It is the risk of one failure cascading everywhere. The tool forces you to define bulkheads and blast radius limits so that a single shard failure doesn't take down your entire system. **Is this for high-level project planning?** No, it is for deep operational readiness. It focuses on capacity, cost, and reliability metrics. It is designed to catch fatal gaps in system design before they hit production. **How do I know if my plan is actually good?** The Connector provides a Verdict Matrix that names the exact operational axis where your plan is currently failing. This tells you exactly what to fix to reach a proven status. **Why does it reject 'we will scale as needed'?** 'We will scale' is hope, not a capacity model. A wartime COO demands numbers: arrival rate (1,200 req/s), service rate (1,800 req/s across 6 workers), utilization (67%), and queue drain behavior under burst (200ms p99). Without these numbers, you do not know when the system saturates. 'Auto-scale' is not modeling — it is outsourcing the thinking to the cloud provider. **Why does it demand cost at 3 data points?** Because 'economies of scale' is a claim — not proof. Anyone can say costs decrease at scale. Proof means showing per-unit cost at 3 specific volume points: '$0.12/user at 10K, $0.04/user at 100K, $0.008/user at 1M.' And naming the mechanism: shared compute amortization, committed use discounts, CDN cache hit ratio. If the cost per unit stays flat, you have a service agency, not a platform. **What is 'Accountability Theater'?** It is when you write an SLA that says '99.9% uptime' but the consequence for missing it is 'we will investigate.' That is a press release, not accountability. SRE error budgets require automated penalties: if the error budget burns (43.8 minutes/month for 99.9%), feature deploys freeze automatically until reliability recovers. 'Best effort' is the opposite of a mechanism.