Analyze CSV
The Analyze CSV block takes raw CSV data and a natural language analysis request, then processes them together to produce a result.
Analyze CSV data using AI-generated code
The Analyze CSV block takes raw CSV data and a natural language analysis request, then processes them together to produce a result. It accepts two string inputs: the CSV content itself and a prompt describing what analysis you want performed on that data. The block runs the analysis using code and returns the output through a single result variable.
The key inputs are csvData, which holds the CSV content as a plain string, and analysisPrompt, which is a free-text question or instruction describing what you want to know or extract from the data. The block outputs a result variable containing the analysis output, which can then be passed to downstream blocks in your workflow.
This block fits into workflows where structured tabular data needs to be interpreted, summarized, or queried without requiring manual scripting. It is useful in pipelines that ingest uploaded files, pull data from external sources, or process reports, and need to derive insights or answers from that data programmatically.
What you can build
Real-world workflows powered by the Analyze CSV block.
Sales Report Summarization
Pass a monthly sales CSV and ask the block to summarize total revenue by region, feeding the result into a report generation step.
Data Quality Checks
Submit a dataset CSV and prompt the block to identify missing values or duplicate rows before the data moves to a processing stage.
Customer Segmentation Analysis
Provide a customer export CSV and request a breakdown of segments by purchase frequency or demographic field.
Financial Anomaly Detection
Feed transaction records as CSV and ask the block to flag entries that fall outside expected value ranges.
Survey Response Aggregation
Input survey results as a CSV string and prompt the block to count responses per answer option across all questions.
Inventory Level Monitoring
Pass an inventory CSV and ask the block which items are below a reorder threshold, routing the result to a notification workflow.
Ready to add Analyze CSV to your workflow?
Get Started FreeCommon questions about Analyze CSV
What input parameters does the Analyze CSV block require?
The block requires two string inputs: csvData, which is the CSV content passed as a plain string, and analysisPrompt, which is a natural language description of what analysis you want performed on that data.
What does the block return?
The block returns a single output variable called result, which contains the output of the analysis performed on the provided CSV data.
What kinds of workflows is this block commonly used in?
It fits into workflows that ingest tabular data from file uploads, external data sources, or prior workflow steps and need to extract insights, answer questions, or summarize the data before passing results downstream.
Does the CSV data need to be formatted in a specific way?
The csvData input is typed as a string, so the CSV content should be passed as a plain text string. The metadata does not specify additional formatting constraints beyond that.
Can the analysis prompt be any natural language question?
Yes, the analysisPrompt input accepts any string describing what you want to know about the CSV, such as a summary request, a filtering condition, or a calculation instruction.
Related capabilities
Transcribe Audio
Convert an audio file to text using a transcription model.
Analyze Image
Analyze an image using a vision model based on a text prompt.
Analyze Video
Analyze a video using a video analysis model based on a text prompt.
Deep Research
Sourced information from the web
OCR Image
Extract text from an image
Add Analyze CSV to your workflow
Build powerful AI workflows with drag-and-drop blocks. No coding required.