skills.rest
Datus-ai
Datus-aiOFFICIAL
Explore all skills by this author →
WebsiteGitHub

osi-metrics-authoring

Authors OSI core semantic model metrics from SQL queries and natural language.

1.7k|248|Updated Jul 4, 2025
Search Tags:#metrics#sql#window functions#period-over-period#osi#Semantic Model#Semantic Layer
One-click install
npx skills add https://github.com/Datus-ai/Datus-agent --skill osi-metrics-authoring
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: osi-metrics-authoring
Source: https://github.com/Datus-ai/Datus-agent/tree/main/datus/resources/skills/osi-metrics-authoring
Command: npx skills add https://github.com/Datus-ai/Datus-agent --skill osi-metrics-authoring

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Defining reusable business metrics in a semantic model is error-prone: engineers must translate SQL aggregates, window functions, and period-over-period logic into a consistent OSI core schema without leaking execution-engine syntax. This Skill standardizes that translation so metrics are created, updated, or deleted correctly inside an existing OSI semantic model.

Core Features & Use Cases

  • OSI Metric Authoring: Creates and updates metrics with qualified aggregate expressions, descriptions, ai_context instructions, and DATUS custom_extensions hints such as time_dimension, subject_path, format, and unit.
  • Window and Period-over-Period Semantics: Converts rolling, cumulative, and LAG-based SQL into standalone metrics using window, grain_to_date, window_aggregation, and period_over_period extension fields.
  • Filter Classification and Skip Gates: Distinguishes durable metric conditions from query-time filters, and skips non-metric outputs like rankings or detail lists with a structured skipped status.
  • Use Case: Given a SQL query computing monthly year-over-year revenue growth, the Skill binds the existing semantic model, publishes a revenue_month_yoy metric with a period_over_period extension, validates the model, and publishes it to the knowledge base.

Quick Start

Ask the agent to create a monthly year-over-year revenue growth metric from this SQL query in the existing OSI semantic model.

Frequently Asked Questions about osi-metrics-authoring

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

FAQPage Schema
How do I create a metric in an OSI semantic model from SQL?▼

Bind the existing semantic model target, then write the aggregate business expression with dataset-qualified columns and call upsert_osi_metrics with the complete metric objects. After validation passes with validate_semantic, call publish_metrics to sync the knowledge base.

How do I author rolling or year-over-year metrics in OSI?▼

Write the base aggregate as the OSI expression and encode window, grain_to_date, window_aggregation, or period_over_period semantics in the DATUS custom_extensions JSON. Each window or offset variant is a new standalone metric, never a reference to its base metric.

Does this Skill write MetricFlow YAML or measure definitions?▼

No. It authors OSI core semantics only and never writes MetricFlow YAML, measure_proxy, type_params, or ratio blocks. The Datus OSI compiler infers backing measures and lowers the model to the execution engine.

When should a SQL WHERE clause become part of a metric?▼

A predicate becomes a metric condition only when durable intent is confirmed, such as an explicitly requested named filtered KPI or an established business definition. Otherwise it stays a query-time filter and the base metric is reused.

What happens if the target semantic model does not exist yet?▼

The Skill stops and reports that gen_semantic_model must run first, since it only edits metrics and narrowly required datasets within an existing model. It does not create models, relationships, or model metadata from scratch.

Related Skills

Explore More Data & Analytics Skills→
slingr-stackslingr-stack

backend-datamodels

Define and validate Drumr Framework backend data models with fields, relationships, and permissions.

Official
Advanced
rritecrritec
3

fabriciq

Answer natural-language questions about Power BI report data by querying semantic models.

Community
Advanced
Datus-aiDatus-ai
1.5k

gen-metrics

Convert natural language metric descriptions and SQL queries into validated MetricFlow definitions.

Official
Advanced
Datus-aiDatus-ai
1.5k

metricflow-semantic-authoring

Generate validated MetricFlow semantic model YAML from database table schemas.

Official
Advanced