verify

Compare live mart values to predefined verifications and classify results by status.

1|1|Updated Mar 26, 2026

Search Tags:#validation #dbt #pipeline #verify #data-quality #dive #marts

Ready to run

npx skills add https://github.com/Soria-Inc/soria-stack --skill verify-soria-inc

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: verify
Source: https://github.com/Soria-Inc/soria-stack/tree/main/verify
Command: npx skills add https://github.com/Soria-Inc/soria-stack --skill verify-soria-inc

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Prove data correctness by building a case with evidence across the data pipeline, comparing live mart values to configured verifications.

Core Features & Use Cases

Tiered verification across ingest, dive, and marts to detect anomalies and root-cause data quality issues.
Automated cell-level checks against verifications.csv, plus guidance for remediation and gap analysis.
Use case: after a new ingest, verify checks highlight any regressions and trigger targeted reprocessing or investigation.

Quick Start

Run a verification cycle against the naic_national_kpis_by_company model and review the results.

Frequently Asked Questions about verify

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

FAQPage Schema

How do I verify data quality across a dbt pipeline and detect mart regressions?▼

Verify data quality across a dbt pipeline by comparing live mart values against predefined checks in a verifications seed. This classifies results by status to detect anomalies and highlight regressions after new data ingestion.

What is the best way to automate cell-level data validation for post-ingest checks?▼

Automate cell-level data validation by running checks against a configured verifications.csv file. This surfaces data quality issues across post-ingest and post-dive stages, providing guidance for targeted remediation and gap analysis.

How do I run a verification cycle against a specific marts model?▼

Run a verification cycle by querying actual mart data for a specific model, such as naic_national_kpis_by_company, and comparing the live values to predefined verifications to classify the results by status.

Do I need a verifications seed to check data quality across marts models?▼

Yes, you need access to a verifications seed to check data quality across marts models. The Skill requires this seed and the ability to query actual mart data to classify verification results and identify anomalies.

Can I use this for ongoing data governance or is it only for post-ingest checks?▼

You can use this for ongoing data governance across any marts model with configured verifications. It is applicable for both post-ingest and post-dive checks, supporting tiered verification to root-cause data quality issues.

Why does data quality verification fail to surface anomalies in my pipeline?▼

Data quality verification may fail to surface anomalies if the verifications seed is not configured correctly or if it cannot query the actual mart data. It requires both the seed definitions and live mart access to classify results by status.