skills.rest
openai
openaiOFFICIAL
Explore all skills by this author →
WebsiteGitHub

create-data-context

Creates and maintains source-backed semantic-layer skills for data analytics workflows.

5.3k|765|Updated Mar 4, 2026
Search Tags:#data analytics#data warehouse#Automation#data context#metric definitions#source inventory#Semantic Layer
One-click install
npx skills add https://github.com/openai/plugins --skill create-data-context
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: create-data-context
Source: https://github.com/openai/plugins/tree/main/plugins/data-analytics/skills/create-data-context
Command: npx skills add https://github.com/openai/plugins --skill create-data-context

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

Data analysts repeatedly re-explain metric definitions, canonical tables, and dashboard caveats to AI assistants. This Skill captures that knowledge once as an inspectable, citable semantic-layer skill so future data analysis starts from shared, source-backed context.

Core Features & Use Cases

  • Semantic Layer Creation: Crawls warehouses, dashboards, SQL, docs, repos, and team channels to build a compact semantic-layer skill with metric definitions, table grain, joins, and caveats.
  • Runtime-Aware Persistence: Installs the generated skill into ChatGPT personal Skills, ChatGPT Desktop skill directories, or returns a portable package when no persistent destination exists.
  • Weekly Refresh Automation: Offers an optional scheduled polling automation that checks the source inventory for changes and proposes validated updates.
  • Use Case: A data team points the Skill at their dbt repo, a verified Looker dashboard, and a Slack channel; it produces a payments-semantic-layer skill that future analyses cite for canonical revenue definitions.

Quick Start

Ask the assistant to create a semantic layer for your product area using your metric docs, dashboards, and warehouse tables as sources.

Frequently Asked Questions about create-data-context

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

FAQPage Schema
How do I create a semantic layer for my data team?▼

Provide a target product or business area plus starting sources such as transformation code, metric docs, verified dashboards, or warehouse tables. The Skill crawls those sources and generates a compact semantic-layer skill with references for metrics, tables, query patterns, and caveats.

What sources can be used to build a data semantic layer?▼

Supported source lanes include data warehouses like Databricks, BigQuery, and Snowflake, BI dashboards, raw SQL, data documentation, code repositories, notebooks, and team communication channels like Slack. Durable sources such as transformation code and verified dashboards take precedence over query history or chat discussion.

Where is the generated semantic-layer skill saved?▼

The destination depends on the runtime: ChatGPT web installs to personal Skills, ChatGPT Desktop writes to $CODEX_HOME/skills/<area>-semantic-layer, and unsupported runtimes receive a portable package. The Skill reports the exact path or installed link after creation.

Can the semantic layer stay updated automatically?▼

Yes, after creation the Skill offers an optional weekly polling automation that checks the source inventory for changes and proposes validated updates. It is only created with explicit user approval and requires a stable skill path plus a usable source inventory.

When should I not use this skill for data work?▼

Do not use it for ordinary data tasks like answering a metric question or building a report; those route to the Data Analytics index skill. Use it only when you want to save, update, inspect, or repair reusable data context as a semantic layer.

Related Skills

Explore More Data & Analytics Skills→
tomtoto757tomtoto757
14

financial-analytics-dashboard

Create interactive financial KPI dashboards with drill-downs and threshold alerts.

Community
Intermediate
metabasemetabase
49.0k

metabase-data-app-semantic-layer

Build React data apps that query Metabase tables and metrics through generated TypeScript schema files.

Official
Advanced
lookleelooklee

Loop - 循环任务调度器

Schedule recurring commands with cron-like logic and immediate execution.

Community
Intermediate
RemseyMailjardRemseyMailjard

wiki-qa

Analyze code repositories to answer questions with source-file citations.

Community
Advanced