topic Re: Setup Qlik Cloud MCP & LLM Copilot in Data Movement & Streaming

Setup Qlik Cloud MCP & LLM Copilot

FelipeD — Thu, 19 Mar 2026 20:51:38 GMT

Hi team,

I tried to set up this LLM but currently, to make the test with Swagger on Power Automate to authorize the OAuth session, it indicates that the redirect URI is not registered, and I made this in the before steps.

Some have the same problem or know other steps to work with Qlik Cloud MCP and Copilot.

Thanks

Re: Setup Qlik Cloud MCP & LLM Copilot

avinashelite — Fri, 20 Mar 2026 07:18:51 GMT

Check this

https://help.qlik.com/en-US/cloud-services/Subsystems/Hub/Content/Sense_Hub/QlikMCP/Administering-Qlik-MCP.htm

Re: Setup Qlik Cloud MCP & LLM Copilot

JacovCohenQ — Sun, 05 Apr 2026 12:01:08 GMT

Has anyone managed to create such a working connection between Qlik MCP and Copilot?

Re: Setup Qlik Cloud MCP & LLM Copilot

marksouzacosta — Mon, 06 Apr 2026 02:04:45 GMT

From my research:

Right now, Microsoft Copilot doesn’t expose a feature that lets you attach or register your own MCP servers. That means:

You can’t add custom MCP tools
You can’t point Copilot at your own MCP server
You can’t extend Copilot’s capabilities through MCP the way you might with other platforms

However, if you are a user of Microsoft Copilot Studio, you can use MCP Servers:
https://learn.microsoft.com/en-us/microsoft-copilot-studio/agent-extend-action-mcp

I tried to setup the Microsoft Copilot Studio, but I could not access the MCP Server part. Maybe the trial account does not allow that.

Regards,

Mark Costa

Re: Setup Qlik Cloud MCP & LLM Copilot

nicodataiq — Tue, 07 Apr 2026 14:42:29 GMT

Hi! I managed to get this working without issues. Microsoft Copilot Studio allows you to add custom MCP Servers to your agents, however, there is a bug to be aware of: tools that require a "measure" as input don't expose that parameter properly, which means the agent in Copilot Studio won't be able to use those tools as expected.

If you're looking to integrate with Copilot Studio, I'd be happy to share a mini tutorial with the steps I followed. Just let me know!

Re: Setup Qlik Cloud MCP & LLM Copilot

jomar_ebonite — Wed, 08 Apr 2026 08:02:19 GMT

Hi @nicodataiq if you could share a tutorial on how you setup this, that would be a great help. appreate it!

Re: Setup Qlik Cloud MCP & LLM Copilot

banderson735 — Wed, 08 Apr 2026 16:45:24 GMT

@FelipeD - I just posted a community article on how to use Microsoft's Agent Toolkit to create a Copilot agent that integrates with Qlik's MCP Server.

Re: Setup Qlik Cloud MCP & LLM Copilot

hugo_andrade — Wed, 08 Apr 2026 18:31:25 GMT

Based on the replies in this thread, the bigger point is to first confirm which Microsoft product you are using. A few people noted that regular Microsoft Copilot is not the same as Copilot Studio for MCP scenarios, and another member said they got it working in Copilot Studio, with one limitation around tools that require a measure parameter. There is also now a Qlik article shared by @banderson735 showing an Agent Toolkit approach for integrating a Copilot agent with Qlik MCP.

My first step would be to follow that new article from @banderson735 and test against Copilot Studio / Agent Toolkit rather than Swagger or standard Copilot first. That should help separate a true OAuth redirect problem from using the wrong Microsoft entry point for MCP.

Hope this helps. Please give it a like, mark it as Helpful or Solution if that was the case.

Re: Setup Qlik Cloud MCP & LLM Copilot

nicodataiq — Thu, 09 Apr 2026 13:20:22 GMT

Hi everyone! As promised, here's a quick tutorial on how I got Qlik Cloud MCP working with Microsoft Copilot Studio.

Tutorial: Connecting Qlik MCP with Copilot Studio

1. Create the OAuth Client in Qlik Cloud

Go to Qlik Cloud → Administration Activity Center → OAuth → Create new

Configure it as follows:
- Client type: `Web`
- Scopes: `user_default`, `mcp:execute`
- Redirect URLs: leave empty for now

Save and copy your Client ID and Client Secret.

> ⚠️ Important: Make sure to select **Web** as the client type, if you choose *Native*, no Client Secret is generated and the integration will not work.

2. Create the MCP Tool in Copilot Studio

Go to Tools → Add Tool → Model Context Protocol → Manual and fill in the fields:

Replace `<your-tenant>` with your actual Qlik Cloud domain, then save.

3. Complete the Redirect URL back in Qlik

After saving in Copilot Studio, a Redirect URL is automatically generated.

Go back to Qlik → OAuth → Edit the client you created → paste that Redirect URL exactly as it appears → Save.

That's it! Hope this helps someone. Happy to answer questions if anything is unclear. 🙌

⚠️ One open issue: I'm still hitting a bug with the input fields on tools that require measures, if anyone has figured out how to work around that, I'd love to hear from you!

Re: Setup Qlik Cloud MCP & LLM Copilot

paulcalvet — Sun, 12 Apr 2026 08:53:05 GMT

Hello,

Many thanks for this quick tutorial.

I try to configure it and I have an error when I click on the buton to retrieve the qlik capabilities.

I will recreate it with your tutorial.

What do you mean when you said : I'm still hitting a bug with the input fields on tools that require measures, if anyone has figured out how to work around that, I'd love to hear from you!

Thanks !

Re: Setup Qlik Cloud MCP & LLM Copilot

paulcalvet — Mon, 13 Apr 2026 13:26:47 GMT

I try it, and Copilot with Qlik doesn't work very well.

I have the error OpenAIModelTokenLimit, I ask to Claude why and this is the the summary :

Is somebody use it or is everybody use Claude ?

🔴 The OpenAIModelTokenLimit Error — What's Happening

This error means your conversation has exceeded the maximum token limit allowed by the underlying OpenAI model in your Copilot Studio agent.

The real issue is structural: with an MCP like Qlik, every tool call returns data (schemas, query results, metadata) that accumulates in the conversation context — and that fills up the counter very quickly.

📊 Context Window Comparison

Copilot Studio Claude (direct)

Effective context window	~4k–8k tokens (GPT-3.5/4 depending on config)	200k tokens (Sonnet 4.6)
Underlying model	GPT-3.5 / GPT-4 / GPT-4 Turbo depending on plan	Native Claude Sonnet 4.6
Max MCP payload	Limited — if a real-time connector returns a payload larger than the agent can handle, it throws a blocking error	Up to ~180k tokens of data per exchange
History management	Accumulates → hits limit fast	Natively managed, sliding window
Stateful MCP support	Known issue: when an MCP server holds state, errors appear silently on the Copilot Studio backend — the workaround is to build stateless MCP servers	No restriction on stateful servers
Context control	Very limited for end users	Fully controllable

⚠️ Copilot Studio–Specific Limitations with Qlik MCP

1. Token limit too low for Qlik data Qlik MCP can return large JSON objects (datasets, hypercubes, app lists). With GPT-4 at 8,192 tokens, just a few exchanges are enough to saturate the context.

2. Statefulness problem Copilot Studio creates a session ID each time it interacts with a stateful MCP server — this causes silent backend errors. The current recommendation is to build stateless MCP servers for Copilot Studio. A Qlik MCP that maintains an active connection is typically stateful by nature.

3. Connector payload limits If a real-time connector returns a payload larger than the agent can handle, the error is blocking with no graceful fallback.

4. No granular control Unlike Claude directly, you can't easily adjust what enters the context, truncate MCP responses, or manage history compression yourself.

✅ Why Claude Is Better Suited for Qlik MCP

200k token effective context: you can pass full Qlik hypercubes, app metadata, and multiple exchanges without saturating
No intermediary layer: Claude calls the MCP directly, without Copilot Studio's constraints (OpenAPI 2.0 requirement, payload limits, etc.)
Native long-context handling: ideal for multi-step analytical sessions querying Qlik iteratively
No GPT model dependency: you avoid OpenAIModelTokenLimit errors by design

🔧 Workarounds If You Must Keep Copilot Studio

Reduce the Qlik payload — configure the MCP to return paginated or truncated responses
Reset the conversation regularly (new session) to clear the accumulated context
Switch to GPT-4 Turbo in your Copilot Studio settings (128k tokens) if your license allows it
Offload heavy lifting to Power Automate — aggregate and summarize Qlik data in a Flow before sending a compact result to the agent

Bottom line: for any serious use of Qlik MCP (iterative queries, large metadata volumes, multi-step analysis), Claude is structurally better suited than Copilot Studio, whose token limits are simply too low for this kind of workload.

Re: Setup Qlik Cloud MCP & LLM Copilot

nicodataiq — Tue, 14 Apr 2026 18:56:40 GMT

Replying to the bug question:

So basically MCP tools expose their arguments so the agent knows what to send. The problem is that the "measures" argument isn't being exposed properly, so Copilot ends up calling the tool with everything except measures and it throws an error. Haven't cracked it yet, if you find something let me know!

Re: Setup Qlik Cloud MCP & LLM Copilot

nicodataiq — Tue, 14 Apr 2026 18:58:22 GMT

Replying to your last post:

I also experienced issues when using GPT models in Copilot, especially with errors that block access to sensitive information and prevent retrieving table data. You can solve this issue by using Claude models.

In my experience, it’s still possible to make an agent work with some effort, particularly for analytics on existing Qlik charts. You can improve results through better prompting and by adding automation flows to the agent. For example, you can retrieve large JSON datasets, process or filter them with a flow, and then return a more compact and usable result back to the agent.