What problem does it solve? Setting up a dlt filesystem pipeline involves many decisions—backend credentials, file readers, glob patterns, table naming, and secrets configuration—that are easy to get wrong on the first attempt. This Skill scaffolds and configures a working filesystem pipeline step by step, so data flows from files into a destination without trial-and-error debugging. ## Core Features & Use Cases - Guided scaffolding: Runs dlthub pipeline init in the working directory, then replaces the demo code with a focused single-table or multi-table pipeline function. - Multi-backend support: Handles Local disk, S3, GCS, Azure, and SFTP with the correct dlt extras and per-backend credential placeholders written via MCP tools. - Safe first runs: Enumerates matched files before loading, recommends sample runs with add_limit(1), and provides a debug table for common first-run errors. - Use Case: You have a folder of CSV reports in an S3 bucket and want them in DuckDB. The Skill collects your destination, backend, and bucket URL, scaffolds the pipeline, writes placeholder credentials, and walks you through a verified sample load. ## Quick Start Ask the assistant to create a dlt filesystem pipeline that loads CSV files from your S3 bucket into DuckDB, providing your bucket URL when prompted.
