Steve-t-hill
- 5 skills
- 0 followers
- 7 hours ago last updated
- ▌ Bigquery GRAPHQL · steve-t-hill bundleExpert guidance and automated agent protocol for building BigQuery Labeled Property Graphs (LPG), querying with GQL (ISO standard), executing a 5-step relational-to-graph translation (Golden Queries, Topology Extraction, ML.DESCRIBE_DATA profiling, Logic Mapping, DDL Generation), optimizing multi-hop queries, and implementing graph measures. Use this skill when: 1. Translating existing BigQuery relational tables (99% of cases) or unstructured documents (1% of cases) into CREATE PROPERTY GRAPH DDL. 2. Executing automated agent pipelines for topology extraction (via Data Catalog MCP) and statistical profiling (ML.DESCRIBE_DATA) to fold lookup tables into properties and discover hidden FKs. 3. Writing high-performance GQL queries, handling super nodes, using VECTOR_SEARCH with graph traversals, or querying graph measures via GRAPH_EXPAND and AGG().
- ▌ Bigquery Ca Builder · steve-t-hill bundleExpert guidance and automated agent protocol for generating configuration and deployment artifacts for BigQuery Conversational Analytics (BQCA) Agents. This skill automates the creation of 'config.yaml', system instructions, agent-level glossaries, table schema metadata, join relationships, and Python-based synchronization scripts to enable NL2SQL and GQL capabilities against existing datasets and Property Graphs. Use this skill when: 1. Building a new conversational analytics agent from scratch. 2. Generating or updating system instructions, agent-level glossaries, or table relationships for an NL2SQL agent. 3. Scaffolding deployment scripts for BQCA lifecycle management.
- ▌ Kc Data Domain Builder · steve-t-hill bundleUse this skill to create, manage, bind, and inspect Dataplex Data Domains in Google Cloud Knowledge Catalog. Use when the user wants to organize data assets logically by business structure, authorize domains, or link BigQuery/GCS resources to a domain. Do not use for general Dataplex catalog search outside of domain contexts, or for managing "Data Products" (which package assets for distribution rather than logical ownership).
- ▌ Kc Data Product Builder · steve-t-hill bundleUse this skill to create, manage, bundle, and secure Data Products in Google Cloud Knowledge Catalog. Use when the user wants to package related data assets (BigQuery tables, views, GCS buckets) into curated bundles designed to solve specific business use cases or streamline access for consumers. Do not use for logical ownership organization (which is handled by Data Domains) or general catalog browsing.
- ▌ Kc Business Glossary Builder · steve-t-hill bundleExpert guidance and automated agent protocol for managing Google Cloud Knowledge Catalog Business Glossaries. This skill provides the ability to create, update, and manage glossaries, categories, and terms, as well as link terms to data assets and columns. It enforces a strict "Exhaustive Analysis" rule to ensure 100% conceptual coverage of BigQuery datasets in a single pass. Use this skill when: 1. Building a new business glossary from scratch for a BigQuery dataset. 2. Auditing existing glossaries for missing terms or technical mappings. 3. Establishing data stewardship and ownership metadata across a business ontology. 4. Linking related or synonymous business concepts to improve discoverability.