Document Dataset

---

name: document-dataset description: Generate a comprehensive markdown documentation card for a Dremio dataset user_invocable: true triggers:

• "document dataset"
• "describe table"
• "dataset documentation"
• "data dictionary"
• "what is this table"
• "table info"

---

Document Dataset

Generate a complete documentation card for a Dremio dataset including schema, lineage, sample data, and metadata.

Agent Rules

• Use --fields to reduce output size (e.g., drs schema describe --fields columns.name,columns.type)
• Use drs describe to check parameter schemas before calling unfamiliar commands
• Paths must be dot-separated; quote components containing dots: "my.source".table

Step 1: Get Column Schema

drs schema describe 

This returns:

• Column names and data types
• Nullable flags
• Nested/complex type structures (LIST, STRUCT, MAP)

Record every column — this forms the core of the documentation card.

Step 2: Get Lineage

drs schema lineage 

This shows:

• Upstream dependencies — source tables and views this dataset reads from
• Downstream dependents — views and reflections built on top of this dataset
• Transformation chain — the path from raw sources to this dataset

Step 3: Get Sample Data

drs schema sample 

This returns a small set of representative rows. Use this to:

• Understand actual data formats (date patterns, enum values, ID formats)
• Identify columns that are mostly null or have unexpected values
• Show realistic examples in the documentation

Step 4: Get Wiki and Tags

drs schema wiki 

This returns:

• Wiki content — user-authored descriptions attached to the dataset in Dremio
• Tags/labels — categorization tags applied in the Dremio catalog

Step 5: Check Reflections

drs reflect list 

Document any reflections:

• Reflection type (raw vs aggregate)
• Status (active, refreshing, failed)
• Columns included
• Refresh schedule

Step 6: Assemble the Documentation Card

Combine all gathered information into a markdown card with this structure:

# 

**Path:** ``
**Source:** 
**Tags:** 

## Description

## Schema

| Column | Type | Nullable | Description |
|--------|------|----------|-------------|
|  |  |  |  |
|  |  |  |  |

## Sample Data

## Lineage

**Upstream:** 
**Downstream:** 

## Reflections

| Name | Type | Status | Columns |
|------|------|--------|---------|
|  | raw/agg | active |  |

## Notes

- 

Tips

• For nested types (STRUCT, LIST), expand sub-fields in the schema table with dot notation: address.city, address.zip.
• For large tables (100+ columns), group columns by prefix or functional area.
• If sample data reveals PII (emails, phone numbers, SSNs), note this in the documentation and flag for the data owner.
• Cross-reference with drs access grants if the user wants access info included.

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

• Author: dremio
• Source: dremio/cli
• License: Apache-2.0

Install and usage instructions live in the source repository linked above.