discovering-gcp-data-assets

Discovers and inspects Google Cloud data assets using Dataplex search and context lookup.

1|Updated Jul 4, 2026
One-click install
npx skills add https://github.com/trungenglish/SHOPWISE --skill discovering-gcp-data-assets-trungenglish
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: discovering-gcp-data-assets
Source: https://github.com/trungenglish/SHOPWISE/tree/main/.agents/skills/discovering-gcp-data-assets
Command: npx skills add https://github.com/trungenglish/SHOPWISE --skill discovering-gcp-data-assets-trungenglish

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Locating the right data asset in Google Cloud is difficult when you only have a keyword or topic, and retrieving schema, metadata, or governance context for BigQuery, Spanner, BigLake, or Cloud Storage assets often requires knowing exact resource IDs upfront. ## Core Features & Use Cases - Semantic and Keyword Search: Query the Dataplex Universal Catalog with natural language intent or precise technical filters like name:, system=, location=, and fully_qualified_name=. - Deep Metadata Lookup: Fetch full schema and business context for any asset using the Dataplex Lookup Context command with full entry names. - Guided Fallbacks: Handle public datasets, missing results, and permission errors with defined fallback paths to native tools like bq CLI. - Use Case: A user asks to find "Q4 product sales data" in BigQuery. The skill runs a semantic Dataplex search, resolves the full entry name, and returns the table's schema and governance metadata. ## Quick Start Find the BigQuery table containing customer order data in my GCP project and show me its schema and metadata.

Frequently Asked Questions about discovering-gcp-data-assets

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I find BigQuery tables in Google Cloud without knowing the table name?▼

Use Dataplex search with a semantic query describing the data's meaning, such as "Q4 product sales data", or a keyword query like `name:order`. The search returns full entry names that you then pass to the context lookup command for schema details.

What is the difference between semantic and keyword search in Dataplex?▼

Semantic search accepts natural language intent and handles synonyms and plurals automatically, requiring uppercase AND/OR operators. Keyword search matches exact technical strings, is case-insensitive for operators, and requires singular terms like "product" instead of "products".

Why does Dataplex context lookup fail with NOT_FOUND?▼

Lookup fails when you pass short table IDs instead of full entry names starting with `projects/`, or when the asset's project has not synchronized metadata with the Universal Catalog. Fall back to native tools like `bq show` for direct inspection.

Can I inspect bigquery-public-data datasets with Dataplex?▼

No, Dataplex Lookup Context fails for assets in the bigquery-public-data project. Inspect these public tables directly using the `bq` CLI or BigQuery MCP tools instead of the search and lookup workflow.

What permissions are required for Dataplex entry search?▼

The project passed via `--project` must have the Dataplex API enabled, and the user must hold the dataplex.entries.get permission. Without these, searches fail with PERMISSION_DENIED errors.

What should I do when Dataplex search returns no results?▼

Try at most three query variations, such as switching to singular keywords, removing projectid or location filters, or using fully_qualified_name. If still empty, stop and ask the user for the project ID, dataset ID, or full entry name.