How does this MCP decide which capability is best?

It uses a deterministic engine to balance the execution cost in milliseconds against your required accuracy thresholds.

Which AI clients can use this MCP?

You can use this with any MCP-compatible client, including Claude, Cursor, Windsurf, and VS Code.

Does this MCP host the capabilities itself?

No, this MCP is a decision engine. It evaluates the capabilities you already have available to help your agent choose between them.

Can I set a minimum accuracy requirement?

Yes, you can use the capabilities to filter for only those variants that meet or exceed your specific accuracy needs.

Is there a cost for using this MCP?

Access is managed through your Vinkius subscription, which includes hosting and management.

How does the capability decide which variant is best?

The engine first filters all variants to find those that meet the min_accuracy requirement. From that subset, it selects the variant with the lowest estimated_cost_ms. If costs are tied, it selects the one with higher accuracy.

Can I see all capabilities that meet my accuracy needs?

Yes, you can use the list_qualified_variants capability to retrieve a list of all available variants that satisfy your minimum accuracy threshold.

What happens if no capability meets the accuracy requirement?

If no capability variants meet the specified min_accuracy, the select_optimal_variant capability will return an error indicating that no suitable variant was found.

An astronaut overseeing a futuristic data hub

Download PDF Markdown Version

Connect Cost-Controlled Tool Selector to ChatGPT and Claude

Set it up once and it works in ChatGPT, Claude, Cursor, and any other client you love. It all runs safely in the background, so you can relax and focus on what really matters. Pick the right capability variant without overspending on accuracy.

Ask AI about this Connector

Developed, maintained, and hosted by Vinkius.

MCP VERIFIED · PRODUCTION READY · VINKIUS GUARANTEED

Waiting for input…

Works with modern AI clients that support MCP, including ChatGPT, Claude, Cursor, and more.

Your AI does the work. We keep it safe and sound.

No other AI tool gives you this. Your workflow runs in the client you already love, your account stays protected, and you decide what the agent can and cannot do. It is the whole reason to make the switch.

Each Connector lives inside our infrastructure, the same place where your workflow runs. The three things below are already part of it, working quietly in the background so you never have to think about them. Nothing to add, nothing to set up, nothing to worry about.

V8 Isolate01

Each Connector runs inside its own sealed environment.

Each Connector runs inside its own sealed V8 runtime — no access to your files, your system, or other servers. A Connector never sees past its own request.

Kill Switch02

One switch stops every agent. Instantly.

One action deactivates every Connector, revokes every token, and ends every session at once. AI Governance keeps the record.

FinOps per User03

Spending limits that pause instead of surprise.

Every user keeps their own connections and credentials, isolated from everyone else. Anything that would cross your spending limit is held for your approval, not billed.

Three controls, built into the layer your AI connects through.

Curated set · 3 capabilities

We ship only the actions that matter.

Capability set01 / 01

01-03

3 capabilities in this set.

Part of 3 available through Cost-Controlled Tool Selector.

01
Compare variant efficiency
This capability calculates the performance-to-cost ratio between two specific capability options. It helps you see which one gives you more accuracy per millisecond of execution.
02
List qualified variants
This capability identifies every capability capable of meeting a specific task requirement. It ignores cost to show you the full list of options that actually work.
03
Select optimal variant
This capability picks the single best capability variant for your specific task. It balances your efficiency needs against your required constraints to find the winner.

Real work

What teams actually run

Dynamic Model Routing

An agent chooses between a small, fast model and a large, slow model based on the complexity of the user prompt.

Budget-Constrained Agents

You limit an agent's total execution cost by forcing it to pick the cheapest capability that still meets a quality bar.

Latency Optimization

The agent selects capabilities that minimize total response time while maintaining necessary accuracy.

A/B Testing Capability Efficiency

You compare two different capability implementations to see which one provides better value for your specific use case.

One connector, every AI

Cost-Controlled Tool Selector works with the most popular AI clients.

These are the most popular clients, each with a step-by-step guide: one link, set up once, with governance and visibility built in. And because everything runs on the MCP standard, the same connection also works in any other compatible client — nothing to rebuild.

Building your own app? The connector is yours to use.

You don't need a client to put Cost-Controlled Tool Selector to work: the same hosted connection plugs into your own applications and agent code, with the same governance on every request. Build with it, chat with it — one connection for both.

Observed, not estimated

857ms average. Fast in production.

Cost-Controlled Tool Selector is checked daily against the live service.

Daily averagePeak 1047ms

Oct 3Today

Fastest day: 727ms
Slowest day: 1047ms
14-day trend: Improving-17%

Connect your client

One URL. Every client.

Activate the Connector, copy your link, and paste it into the client you already use. 3 capabilities arrive ready to run.

Preview access · not provider authentication

The vk_preview_* token belongs to Vinkius preview infrastructure. It lets Claude discover and display the capabilities of Cost-Controlled Tool Selector, so you can see the experience inside your AI.

It does not authenticate your account with Cost-Controlled Tool Selector. Actions requiring credentials or live account data may not run until you activate the Connector and authorize the service.

Cost-Controlled Tool Selector Connector

You're all set. Choose your MCP client and follow the setup instructions.

Connector linkhttps://edge.vinkius.com/vk_preview_EcT3WLB2s90OH9keGSnCvbvlarTyEOUA0zFmootW/mcp

Claude Desktop

Follow the steps below to connect in seconds.

1In Claude Desktop, open Settings → Connectors.
2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
3Click Add and start a new chat — Cost-Controlled Tool Selector capabilities are ready to use.

Configuration · claude_desktop_config.jsonCopy

{
  "mcpServers": {
    "cost-controlled-tool-selector-mcp": {
      "url": "https://edge.vinkius.com/vk_preview_EcT3WLB2s90OH9keGSnCvbvlarTyEOUA0zFmootW/mcp"
    }
  }
}

Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline

Step-by-step instructions for each client are in the guide. How to connect

Guided setup for Claude? Set up Cost-Controlled Tool Selector in Claude and other AIs

See all the AI clients this connector works with ↑

Who it's for

Built for the work Cost-Controlled Tool Selector owners hand off.

This MCP is built for engineers building autonomous agents that need to manage their own resource consumption.

01
AI Engineers
They hand off the decision of which model or capability to call to the agent.
02
LLM Ops Specialists
They use it to keep production inference costs within specific budget limits.
03
Agentic Workflow Developers
They implement it to ensure agents don't waste compute on simple tasks.

FAQ

Questions Cost-Controlled Tool Selector owners ask.

01
How does this MCP decide which capability is best?
It uses a deterministic engine to balance the execution cost in milliseconds against your required accuracy thresholds.
02
Which AI clients can use this MCP?
You can use this with any MCP-compatible client, including Claude, Cursor, Windsurf, and VS Code.
03
Does this MCP host the capabilities itself?
No, this MCP is a decision engine. It evaluates the capabilities you already have available to help your agent choose between them.
04
Can I set a minimum accuracy requirement?
Yes, you can use the capabilities to filter for only those variants that meet or exceed your specific accuracy needs.
05
Is there a cost for using this MCP?
Access is managed through your Vinkius subscription, which includes hosting and management.
06
How does the capability decide which variant is best?
The engine first filters all variants to find those that meet the min_accuracy requirement. From that subset, it selects the variant with the lowest estimated_cost_ms. If costs are tied, it selects the one with higher accuracy.
07
Can I see all capabilities that meet my accuracy needs?
Yes, you can use the list_qualified_variants capability to retrieve a list of all available variants that satisfy your minimum accuracy threshold.
08
What happens if no capability meets the accuracy requirement?
If no capability variants meet the specified min_accuracy, the select_optimal_variant capability will return an error indicating that no suitable variant was found.

Explore

More in Optimization

Connect Cost-Controlled Tool Selector to Claude, Cursor, ChatGPT & more

Your AI does the work. We keep it safe and sound.

Each Connector runs inside its own sealed environment.

One switch stops every agent. Instantly.

Spending limits that pause instead of surprise.

We ship only the actions that matter.

01-03

Compare variant efficiency

List qualified variants

Select optimal variant

What teams actually run

Dynamic Model Routing

Budget-Constrained Agents

Latency Optimization

A/B Testing Capability Efficiency

Building your own app? The connector is yours to use.

Cost-Controlled Tool Selector Connector

Claude Desktop

AI Engineers

LLM Ops Specialists

Agentic Workflow Developers

How does this MCP decide which capability is best?

Which AI clients can use this MCP?

Does this MCP host the capabilities itself?

Can I set a minimum accuracy requirement?

Is there a cost for using this MCP?

How does the capability decide which variant is best?

Can I see all capabilities that meet my accuracy needs?

What happens if no capability meets the accuracy requirement?

Model Routing Efficiency Calculator AI Connector

Mine Ventilation Cost Optimizer AI Connector

Prompt Compression Efficiency Calculator AI Connector

Model Routing Optimizer AI Connector

Suggestions

LLM API Cost Estimator AI Connector

Rice Drying Cost & Optimization AI Connector

Bit Selection Optimization AI Connector

Wine Shipping Cost Optimizer AI Connector