finding-data-lake-assets

What problem does it solve?

This Skill resolves ambiguous data lake or lakehouse asset references into concrete Glue, S3, S3 Tables, and Redshift catalog entries so you can quickly identify the right dataset and where it lives.

Core Features & Use Cases

• Guided asset resolution: Turns a table name, keyword, column name, or S3 path into a specific table reference with format and location where available.
• Layered discovery with early stopping: Searches Glue first for high-confidence matches, then uses S3 reverse lookup or Redshift only when appropriate to minimize latency and token usage.
• Confidence-gated disambiguation: Returns immediately on high confidence, lists top candidates for medium confidence, and reports what was searched or skipped when results are weak or absent.

Use case example: You say “what is our customer_id dataset in us-east-1?” and the Skill finds the best matching Glue table (or S3/S3 Tables/Redshift counterpart when relevant) and returns the resolved reference plus the sources searched/skipped.

Quick Start

Ask: "Resolve customer_id data in us-east-1 and tell me the matching Glue/S3/S3 Tables/Redshift table reference and location."