Unlock a world of possibilities! Login now and discover the exclusive benefits awaiting you.
Improving Declarative Pipeline Development
If you've been building Qlik Talend Cloud pipelines declaratively with a coding agent like Anthropic's Claude Code or OpenAI's Codex, then you already know the productivity win. And if you've coded for any length of time, then you may have encountered a friction point where your agent needed a real connection name, or a real table ID, or a real value from your tenant, and template placeholders just won't cut it. Until now, that meant tabbing out of VS Code, opening the Qlik Cloud web app, and hunting down the answer by hand.
That changes today. Qlik's MCP server has just added three new lookup tools built specifically for declarative data pipeline development, closing that gap by bringing real tenant context directly into your coding session.
New MCP Tools for Agentic Data Engineering
The three new MCP tools are as follows:
- qlik_search: Searches for Qlik catalog resources such as applications, datasets, data products, glossaries, etc. by name or content. This tool has been extended to also return pipeline projects and pipeline connections.
- qlik_get_pipeline_project_details: Returns the project’s binding values. Used to discover all of the details about a specific pipeline project
- qlik_search_connection_objects: List all source entities (tables, views) available in a data connection. Used to discover what data sources are available when configuring data pipeline tasks.
In practice, that means you can ask the tenant a question via your coding agent and get a straight answer: does this space exist, what's this connection actually called, what tables sit behind it. No more generating a pipeline against a guess and finding out it's wrong on import.
So now you can ask questions while you are developing like:
"Show me all available tables in our Sales database connection so I can decide which tables to include in the data pipeline."
And you'll receive context-aware answers as below:
Why it Matters
The benefit is simple: less time spent switching context and more time spent writing pipelines that work the first time. A data engineer working on a pipeline project can ask about spaces, connections, and table structures in natural language right in the editor and keep coding. That's a real efficiency gain for iterative development. It also directly addresses the top piece of feedback we heard during the original testing phase for declarative pipelines: many data engineers wanted a mechanism to introspect a live Qlik environment for information beyond their current pipeline project. So these tools fill that gap. They search, they describe, and they resolve. Delivering precise answers on the current Qlik environment, highlighting what is and isn't with data pipelines.
Available Now
The new tools for pipelines are live as of August 25th and work with any third-party coding agent connected to Qlik's MCP server, including Claude Code, GitHub CoPilot and Open AI's Codex. Note that if you're already building declarative pipelines and MCP is connected, then there's nothing new to set up. The tenant pipeline context is just there the next time you ask a question.
However, if you haven't tried declarative pipelines with a coding agent yet, then now's a good time to start. All the instructions for configuring your IDE, adding pipeline schemas, configuring Git Hub, and setting up MCP for coding agents can be found on the Qlik Developer portal here.
Furthermore, if you've not yet signed up for a Qlik Talend Cloud trial tenant, then you can register for free here too!
Finally, the only thing left for me to say is "Happy coding pipelines with enhanced MCP tools!". We know you're going to love it.