Agentic RAG SupabaseInstall

Overview

The node "Agentic RAG Supabase" provides functionality to process files, vectors, and agentic retrieval-augmented generation (RAG) workflows using a Supabase backend with pgvector support. Specifically for the File resource and Parse File operation, the node reads and parses a file from a given file path, extracting its textual content and splitting it into manageable chunks.

This is useful in scenarios where you want to ingest documents such as PDFs, text files, or Word documents into an automated workflow for further processing like embedding generation, structured data extraction, or vector search. For example, you might use this node to parse a PDF report into text chunks that can then be embedded and stored for semantic search.

Properties

Output

The output JSON object contains the following fields:

• text: The extracted plain text content of the file.
• chunks: An array of text chunks derived from the full text, each chunk being a segment of the text split by word count (default chunk size 200 words with 20 words overlap).
• fileType: The file extension/type detected (e.g., .pdf, .txt, .docx).
• fileName: The base name of the file (filename with extension).

This output allows downstream nodes to access both the raw text and segmented chunks for further processing such as embedding generation or structured extraction.

Dependencies

• Requires access to the local filesystem to read files at the specified path.
• Supports parsing:
  • PDF files via the pdf-parse library.
  • DOCX files via the mammoth library.
  • Plain text files (.txt) via standard UTF-8 reading.
• No external API calls are made during parsing itself.
• Node requires credentials for Supabase and Huggingface inference for other operations but not specifically for parsing.

Troubleshooting

• Unsupported file type error: If the file extension is not .pdf, .txt, or .docx, the node will throw an error stating unsupported file type. Ensure your input file is one of these supported formats.
• File not found or inaccessible: If the file path is incorrect or the node does not have permission to read the file, it will fail. Verify the file path and permissions.
• Corrupted or malformed files: Parsing may fail if the file is corrupted or not properly formatted, especially for PDFs and DOCX files.
• Large files: Very large files may cause performance issues or timeouts depending on environment limits.

Links and References

• pdf-parse npm package — Used for PDF text extraction.
• mammoth npm package — Used for extracting text from DOCX files.
• Supabase pgvector documentation — For vector storage backend.
• Huggingface Inference API — Used for embedding generation in other operations.

This summary focuses exclusively on the File > Parse File operation as requested.