When files are passed directly to an LLM, processing is limited by the model’s own multimodal capabilities. While the Doc Extractor node can convert some file types to text for LLMs to process, its supported formats are still limited.
This is where the Upload File to Sandbox node helps. Once files are uploaded, downstream Agent nodes (with Agent Mode enabled) can execute commands to install tools and run scripts in the sandbox as needed, converting files into formats LLMs can process.
Processing results depends on the model’s reasoning capability.
-
In an Upload File to Sandbox node, select a file variable (e.g., from User Input).
-
In a downstream Agent node’s prompt, reference the
sandbox_path and file_name variables so the model can locate and process the file. For example:
# System instruction
You are a data analyst. Analyze uploaded data files and provide insights.
# User message
Analyze the data in "{{Upload File to Sandbox.file_name}}" at {{Upload File to Sandbox.sandbox_path}}.
The file_name variable includes the file extension (e.g., data.csv).
