MCP Options Order Flow Server: Read Options Flow in Claude and Excel (2026)

MCP options order flow server - if you searched that phrase, you are trying to do one specific thing: point an AI assistant like Claude at live options data so it can tell you where the volume is going, and have those exact numbers available in Excel when you want to dig deeper. This post explains what an MCP server for options order flow actually is, which order-flow primitives the MarketXLS MCP server exposes, and how the same licensed functions that answer a chat prompt also refresh a spreadsheet. It comes with two free Excel workbooks so you can see the flow, not just read about it.

Order flow is the study of what is being traded right now: how many contracts, on which strikes, whether puts or calls are leading, and whether today's activity is far above a name's normal pace. It is context, not a signal, but it is the kind of context that is hard to assemble by hand across a watchlist. An MCP (Model Context Protocol) server makes that assembly a single question away for an AI tool, and MarketXLS ties the same data back into Excel formulas.

MCP Options Order Flow Server: The Quick Comparison

Here is how reading options order flow through an MCP server compares to the two ways most people do it today.

The last row is the point of this guide. When an AI answer and a spreadsheet cell come from the same licensed primitive, calculated the same way, you stop reconciling two versions of the truth and start reading the flow.

What an MCP Options Order Flow Server Does

MCP is an open protocol that lets an AI assistant call external tools in a structured way. An MCP options order flow server is simply an MCP server whose tools return options-flow data: total options volume on an underlying, its trailing average, open interest, put-call ratios, and end-of-day unusual-activity scans.

The MarketXLS MCP server exposes these as callable functions. Ask Claude "which names have unusual options volume today," and it calls the same licensed scan that populates the workbook attached to this post. The value in the chat window and the value in the cell are one number, from one source.

That matters for order flow specifically because flow data is noisy. If your AI is reading one feed and your spreadsheet another, you cannot trust either. A single MCP-backed source removes that gap.

The order-flow primitives

The order-flow story rests on a handful of MarketXLS functions. Every one below was verified against the current function catalog before writing this post.

• =opt_TotalVolumeOptions("NVDA") returns today's total options volume on the underlying.
• =opt_VolumeOptionsAverage("NVDA", 20) returns the 20-day average options volume, the baseline for relative volume.
• =opt_TotalOpenInterestOptions("NVDA") returns total open interest across the chain.
• =opt_PutCallVolRatio("NVDA") returns the put-call volume ratio.
• =opt_PutCallOIRatio("NVDA") returns the put-call open-interest ratio.
• =opt_UnusualStockOptionsActivity(25) returns a spilled table of the day's standout contracts.

These are the same primitives an AI assistant calls through the MCP server. In Excel they are ordinary formulas that recalculate on refresh.

Reading Order Flow: The Metrics That Matter

Order flow only becomes useful once you know which numbers carry information. Four do most of the work.

Relative volume

Absolute options volume is misleading because some names always trade millions of contracts. What matters is today's volume relative to the name's own norm. Divide current volume by its trailing average:

=opt_TotalVolumeOptions("NVDA") / opt_VolumeOptionsAverage("NVDA", 20)

A reading of 2.0x means twice the usual activity. That is the single most useful order-flow tell, because it flags names doing something out of character regardless of their size. The Order Flow Dashboard in the template ranks the whole watchlist by this multiple.

Open interest and the volume-to-OI ratio

Volume tells you how many contracts changed hands; open interest tells you how many positions are still open. When a single strike prints volume well above its open interest, that is the footprint of fresh positioning rather than traders closing out. At the contract level:

=opt_Volume(OptionSymbol("NVDA", DATE(2026,7,17), "Call", 170))
=opt_OpenInterest(OptionSymbol("NVDA", DATE(2026,7,17), "Call", 170))

Divide the first by the second and a value far above 1.0 marks a strike where today's trading dwarfs the standing position. The Strike-Level Flow sheet computes this ratio across the chain so you can see where the day's activity clusters.

Put-call ratios

The put-call volume ratio compares put activity to call activity. Above 1.0 leans defensive or bearish; below 1.0 leans toward call demand. The open-interest version shows the standing tilt rather than today's flow:

=opt_PutCallVolRatio("SPY")
=opt_PutCallOIRatio("SPY")

Read these together with relative volume. A lopsided put-call ratio on a heavy-volume day carries more information than the same ratio on a quiet one.

Unusual activity

The most direct order-flow question is "what is unusual right now." MarketXLS answers it with end-of-day scans that surface contracts trading far outside their norm:

=opt_UnusualStockOptionsActivity(25)
=opt_UnusualOptionVolScanEOD(25)
=opt_UnusualOptionOIScanEOD(25)
=opt_OptionsVolumeLeaders(25)

Each returns a spilled table you can sort and filter. The opt_UnusualStockOptionsActivity function drives the Unusual Activity Scanner sheet in the template.

The Approach: A Repeatable Order-Flow Read

This is an educational framework for organizing what the data shows, not a trading recommendation. Nothing here predicts prices or suggests positions. Options carry risk and are not suitable for every investor.

A disciplined order-flow read moves from broad to narrow:

1. Scan for relative volume. Start on the dashboard and find the names trading well above their average options volume. Those are where something is happening today.

2. Check the put-call tilt. For each flagged name, look at the put-call volume ratio. Is the heavy volume concentrated in calls, puts, or spread across both?

3. Confirm with unusual activity. Cross-reference the unusual-activity scan. Does a specific contract show up with volume far above its open interest? That points to where the positioning sits.

4. Drill into strikes. Open the Strike-Level Flow sheet for the underlying and see whether the flow is clustered at one or two strikes or spread across the chain. Clustered flow at a single strike with a high volume-to-OI ratio is the clearest footprint.

5. Frame it with IV rank. Finally, check where implied volatility sits in its own range with =ImpliedVolatilityRank1y("NVDA"). Heavy flow when IV is already rich means something different than the same flow when IV is depressed.

The workflow is the same whether you ask an AI assistant to run it conversationally or work through the sheets yourself. Because both read the same MCP-backed data, the answers agree.

Building It in Excel with MarketXLS

The order-flow read above is entirely reproducible with MarketXLS formulas. Here is the core of a single-name flow panel.

Underlying context:

=QM_Last("NVDA")                       ' Live underlying price
=ImpliedVolatility30d("NVDA")          ' 30-day implied volatility
=ImpliedVolatilityRank1y("NVDA")       ' Where IV sits in its 1-year range

Volume and open interest:

=opt_TotalVolumeOptions("NVDA")            ' Today's options volume
=opt_VolumeOptionsAverage("NVDA", 20)      ' 20-day average volume
=opt_TotalOpenInterestOptions("NVDA")      ' Total open interest

Positioning:

=opt_PutCallVolRatio("NVDA")           ' Today's put-call volume tilt
=opt_PutCallOIRatio("NVDA")            ' Standing put-call OI tilt

Contract-level flow, built from a constructed option symbol:

=OptionSymbol("NVDA", DATE(2026,7,17), "Call", 170)
=opt_Volume(OptionSymbol("NVDA", DATE(2026,7,17), "Call", 170))
=opt_OpenInterest(OptionSymbol("NVDA", DATE(2026,7,17), "Call", 170))
=opt_ImpliedVolatility(QM_Last("NVDA"), QM_Last(OptionSymbol("NVDA", DATE(2026,7,17), "Call", 170)), DATE(2026,7,17), "Call", 170)
=opt_Delta(QM_Last("NVDA"), QM_Last(OptionSymbol("NVDA", DATE(2026,7,17), "Call", 170)), DATE(2026,7,17), "Call", 170)

OptionSymbol assembles the contract identifier the opt_ functions need, so you never have to hand-type an OCC symbol. Every value recalculates on refresh, which means your flow panel is never stale.

The Template: What Is Inside

The download includes two workbooks. Both carry MarketXLS branding, a "MarketXLS Functions Used" box on every sheet, and links back to the site. Both are built for education, not advice.

Sample workbook - pre-filled with a static snapshot captured on 2026-07-09 so you can explore the layout with no add-in installed. A reference note beside each value shows the MarketXLS formula that would produce it live.

Template workbook - the live version. Every data cell is a MarketXLS formula that updates when you refresh, driven by the yellow input cells on the Inputs sheet.

The eight sheets:

• Cover - overview and table of contents.
• How To Use - a step-by-step walkthrough and the input legend.
• Inputs - the only cells you edit: focus ticker, expiry, the average-volume window, a relative-volume alert threshold, and a put-call threshold. These flow into every other sheet.
• Order Flow Dashboard - the watchlist screener with today's volume, its 20-day average, the relative-volume multiple, open interest, put-call ratios, and IV rank, with conditional formatting that highlights names on alert.
• Unusual Activity Scanner - the day's standout contracts by volume relative to open interest, driven by opt_UnusualStockOptionsActivity, with a plain-language read on each.
• Strike-Level Flow - an at-the-money ladder for the focus ticker showing volume, open interest, the volume-to-OI ratio, IV, and delta at each strike, so you can see where flow concentrates.
• Put-Call Sentiment - the watchlist ranked by put-call ratios, color-coded from call-heavy to put-heavy.
• Methodology and Glossary - data sources, the MCP primitives, and definitions for every term.

Download the templates:

• - Pre-filled with a recent snapshot
• - Live-updating formulas

To make the live version calculate, you need MarketXLS installed in Excel. See the MarketXLS features overview for what the add-in covers, or book a demo to see the options analytics in action.

Reading Order Flow Through an AI Assistant

Because the MarketXLS MCP server exposes these order-flow primitives as callable tools, you can run the entire read conversationally before you ever open Excel. The value of the MCP layer is that the assistant does not guess or summarize from stale training data; it calls the licensed function and reports the number it gets back.

A natural session moves the same way the manual workflow does, just in plain language. You might start broad and let the assistant surface the day's standouts:

• "Which names have the highest relative options volume today across my watchlist?"
• "Show me the unusual options activity leaders and flag whether the flow is in calls or puts."
• "What is the put-call volume ratio on the three names trading above twice their average volume?"

Each of those maps directly to a primitive: relative volume to opt_TotalVolumeOptions over opt_VolumeOptionsAverage, the leaders to opt_UnusualStockOptionsActivity and opt_OptionsVolumeLeaders, and the tilt to opt_PutCallVolRatio. The assistant assembles the answer from live calls, then you move to the workbook to drill into the two or three names worth a closer look.

The handoff is seamless precisely because there is no data mismatch. The relative-volume figure the assistant quotes is the same figure the Order Flow Dashboard shows, because both are the same division of the same two functions. You are not re-checking the AI's work against a separate feed; you are continuing the same read in a different interface. That is what separates an MCP-backed workflow from pasting a screenshot into a chat and hoping the model reads it correctly.

It also keeps the assistant honest about uncertainty. When a scan returns a contract with heavy volume but the direction is ambiguous, the right answer is "this is unusual, here is the volume-to-OI ratio, and the sentiment is not clear-cut," not a fabricated call. Grounding the AI in live functions rather than free-form generation is what makes order-flow answers trustworthy enough to act on your own judgment.

Why One Source for AI and Excel Matters

The reason to run order flow through an MCP server rather than stitching feeds together is consistency. Options data is time-sensitive and easy to misread. If your AI assistant reports one put-call ratio and your spreadsheet shows another, you have to stop and figure out which is right before you can think about the flow itself.

With the MarketXLS MCP options order flow server, that reconciliation disappears. The same opt_PutCallVolRatio call answers a Claude prompt and fills cell C6. The same opt_UnusualStockOptionsActivity scan builds the AI's list and the workbook's scanner. You move between a conversation and a spreadsheet without the numbers shifting, which is exactly what you want when the data changes by the minute.

It also means your workflow scales. Ask the AI a broad question to narrow the field, then open the template to drill into the two or three names worth a closer look. Both halves speak the same language because they read the same data.

Frequently Asked Questions

What is an MCP options order flow server?

An MCP options order flow server is a Model Context Protocol server whose tools return options-flow data, such as volume, open interest, put-call ratios, and unusual-activity scans. It lets an AI assistant like Claude query live options flow. The MarketXLS MCP server exposes these primitives and ties the same functions into Excel, so an AI answer and a spreadsheet cell come from one licensed source.

Which order-flow metrics does it provide?

The core metrics are total options volume (opt_TotalVolumeOptions), average options volume (opt_VolumeOptionsAverage), open interest (opt_TotalOpenInterestOptions), put-call volume and open-interest ratios (opt_PutCallVolRatio, opt_PutCallOIRatio), and end-of-day unusual-activity scans (opt_UnusualStockOptionsActivity, opt_UnusualOptionVolScanEOD). Contract-level volume and open interest are available through opt_Volume and opt_OpenInterest.

How do I read relative options volume?

Relative volume is today's options volume divided by the name's trailing average, for example =opt_TotalVolumeOptions("NVDA")/opt_VolumeOptionsAverage("NVDA",20). A value of 2.0x means twice the normal pace. It is useful because it flags out-of-character activity regardless of a name's usual size. The template's dashboard ranks the whole watchlist by this multiple.

Is unusual options activity a buy or sell signal?

No. Unusual activity, high relative volume, and lopsided put-call ratios are context, not signals. A large print can be a hedge, a spread leg, or a closing trade, and the direction is not always what it appears. This workbook is for education and research only. Nothing in it is investment advice, a recommendation, or a prediction.

Do I need the MarketXLS add-in for the live template?

The live template's formulas calculate through the MarketXLS Excel add-in. Without it, the sample workbook still opens and shows a static snapshot with the formula names noted beside each value, so you can study the structure. To see the numbers update, install MarketXLS and refresh. See the features page for details.

Can I change the tickers and expiry?

Yes. The Inputs sheet holds the watchlist, the focus ticker, the option expiry, the average-volume window, and the alert thresholds. Every other sheet references those cells, so editing them re-points the entire workbook without touching a single formula elsewhere.

The Bottom Line

An MCP options order flow server turns "where is the volume going" from a manual, tab-by-tab chore into a single question your AI assistant can answer, backed by the same licensed data that lives in your spreadsheet. Relative volume, open interest, put-call ratios, and unusual-activity scans are the metrics that carry information, and MarketXLS exposes each of them as both an MCP primitive and an Excel formula. Because both read one source calculated one way, your AI and your workbook never disagree.

Download the two workbooks above to see the flow in a layout you can edit, then explore what the MarketXLS platform covers at marketxls.com or book a demo to see the options order-flow functions live in Excel. This content is educational and is not investment advice; options involve risk and are not suitable for every investor.