Extract ComboCurve Well Data With Claude Via Vinkius Edge

The Manual Data Extraction Trap

If you work in oil and gas, you know the drill. You need to verify a decline curve for a specific well in the Permian. Instead of pulling data directly into your analysis, you log into ComboCurve, navigate through several layers of project folders, find the right forecast, and then… manually export a CSV or take a screenshot.

It is repetitive. It is slow. And more importantly, it creates a gap between having data and actually using it.

When you are evaluating an acquisition or updating a monthly production report, every minute spent clicking through dashboards is a minute lost to actual analysis. The problem isn’t the data; the data is there, sitting in ComboCurve. The problem is the friction of getting that data into your workflow—the one where you are actually making decisions.

Right. So. What if you could just ask?

Introducing the ComboCurve MCP Server

The ComboCurve MCP server changes how you interact with your energy assets. By using the Model Context Protocol (MCP), we have built a bridge between your AI assistants—like Claude Desktop, Cursor, or Windsurf—and the ComboCurve platform.

This isn’t just another API integration. It is an extension of your reasoning engine. Instead of navigating a UI, you use natural language to query your production forecasts, well histories, and project metadata. You are not just retrieving data; you are bringing your data into your conversation.

Through the Vinkius AI Gateway, this connection is managed via Vinkius Edge. This means you do not need to manage complex API environments or handle sensitive credentials directly in your IDE configuration. You use your existing Service Account credentials, and Vinkius handles the routing, authentication, and protection.

Deep Dive into Capabilities

The power of this server lies in its specific toolset. These are not generic functions; they are precise instruments for energy data retrieval.

Project and Well Oversight

You can maintain a continuous view of your assets without ever leaving your coding environment or chat interface. Tools like list_projects and on_project allow you to scan your entire portfolio in seconds. If you need to investigate a specific well, get_well provides the detailed information required for technical audits.

Precision Forecasting

The real value emerges when you query production trends. Using get_monthly_volumes, you can pull monthly forecasted volumes (oil, bbl) for a specific project and forecast.

Imagine running a prompt like this in Claude: “Show me the monthly volumes for the ‘Permian Basin Q1’ forecast.”

The server responds with structured data:

Forecast 'fc-2026' monthly volumes (oil, bbl):
- Jan 2026: 12,400
- Feb 2026: 11,800
- Mar 2026: 11,200
- Apr 2026: 10,650
Decline rate: ~4.8%/month.

No CSV export required. No manual transcription. Just the data you need, exactly when you need it.

Rapid Asset Discovery

If you are performing a rapid acquisition review, search_wells_by_name is indispensable. You can find wells by name or ID instantly, allowing you to build a technical profile of a new lease without the usual dashboard navigation.

Real-World Scenarios

To understand the impact, let’s look at how two different roles actually use this.

Scenario A: The Financial Analyst

A financial analyst is evaluating a new lease acquisition. They need to know if the projected EUR (Estimated Ultimate Recovery) aligns with their valuation model. Instead of requesting a report from the operations team, they use get_scenario_s_details via their AI agent to pull the latest scenario parameters directly into their analysis workflow. The time from inquiry to insight drops from hours to seconds.

Scenario B: The Petroleum Engineer

A reservoir engineer is monitoring decline rates across a specific formation. They notice a potential anomaly in production. Using get_monthly_volumes, they quickly query the monthly volumes for several wells in the Wolfcamp A formation. By seeing the decline curves presented directly in their chat window, they can identify trends and initiate further investigation immediately—without waiting for a weekly report.

Connecting via Vinkius Edge

Setting this up is designed to be frictionless. You do not need to be an integration engineer to make this work.

Through the Vinkius AI Gateway, you use your personal Connection Token, available in your Vinkius dashboard, to link your AI clients. Whether you are using Claude Desktop, Cursor, VS Code, or any MCP-compatible client, the process is the same.

Subscribe to the ComboCurve MCP server in the Vinkius App Catalog.
Configure your client using the Vinkius Edge URL: https://edge.vinkius.com/YOUR_VINKIUS_TOKEN/mcp.
Authenticate using your ComboCurve Service Account credentials (API Key and JWT Bearer Token) when prompted via the secure Vinkius interface.

Vinkius handles the heavy lifting. It manages the authentication, applies rate limits, and ensures that your requests are routed securely to the ComboCurve API. You get the power of an enterprise-grade integration with the simplicity of a single URL.

The Reality Check

No tool is a silver bullet. There are important trade-offs to understand before you deploy this into your workflow.

First, the current implementation is strictly read-only. While you can retrieve well data, project details, and forecasts, you cannot use this MCP server to update well records, modify projects, or change production forecasts. This is a deliberate design choice to ensure that your primary source of truth—the Combo and Curve platform—remains the authoritative record for all write operations.

Second, security is paramount. Because you are connecting an AI agent to sensitive energy data, we have implemented the Security Passport. Every time you use a tool, Vinkius provides transparency into what permissions are being used. You can see exactly what the server is accessing—network access, filesystem, or subprocess execution—and how many tools it exposes.

This architecture ensures that while your AI agent has the data it needs to be useful, you maintain total control over the boundaries of its access.

Conclusion

The era of manual data mining is ending. The transition from reactive reporting—where you are always looking at what happened last week—to proactive, conversational energy management is happening now.

By bringing ComboCurve data into your AI-powered workflows, you are not just saving time; you are changing the fundamental way you interact with your assets. You are moving from clicking and exporting to asking and knowing.

Find the ComboCurve MCP server in the App Catalog and start automating your energy workflows today.