Agent skill
dataasset
Install this agent skill to your Project
npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/other/other/dataasset
SKILL.md
/dataasset
Create a DataAsset note documenting a data entity - database table, API endpoint, data product, Kafka topic, or file. Captures location, ownership, consumers, and planned changes.
Usage
/dataasset <name>
/dataasset "Revenue Fact Table"
/dataasset "Customer Orders"
/dataasset "Maintenance Events"
Instructions
Phase 1: Parse Input & Identify System
- Extract data asset name from input
- Ask which system produces this data:
Which system produces this data? Search: [user searches for System] Or create new System? (Y/n) - Confirm link: "Link to [[System - {{system}}]]? (Y/n)"
Phase 2: Essential Information
Creating DataAsset: {{name}} (produced by {{system}})
1️⃣ Asset ID (unique identifier):
Suggestion: {{SYSTEM}}-{{NAME}}-001
User input: [accept or modify]
2️⃣ Data Type:
- database-table (relational table)
- database-view (virtual table)
- api-endpoint (REST/GraphQL data)
- kafka-topic (event stream)
- data-product (curated dataset)
- data-lake (file-based storage)
- file (CSV, Excel, etc.)
- report (BI report/dashboard)
- cache (Redis/Memcached)
Default: database-table
User input: [selection]
3️⃣ Domain:
- engineering
- data
- operations
- finance
- hr
- supply-chain
- maintenance
Default: [infer from system]
User input: [selection]
4️⃣ Classification:
- public
- internal
- confidential
- secret
Default: internal
User input: [selection]
5️⃣ Storage Location:
Examples: "mydb.fact_revenue", "s3://bucket/path", "/api/v1/orders"
User input: [path/table/endpoint]
6️⃣ Format:
- sql
- json
- parquet
- avro
- csv
- xml
- binary
Default: [infer from data type]
User input: [selection]
Phase 3: Ownership
7️⃣ Data Owner (accountable person):
Search: [[Person - ...]]
User input: [search or skip]
8️⃣ Data Steward (governance contact, optional):
Search: [[Person - ...]]
User input: [search or skip]
Phase 4: Consumer Relationships (Key Differentiator)
Which systems currently consume this data?
Current Consumers (search for Systems):
> [[System - Data Warehouse]]
> [[System - Analytics Platform]]
[Enter blank to finish]
Planned Consumers (systems that WILL consume):
> [[System - New BI Tool]]
[Enter blank to finish]
Deprecating Consumers (systems moving AWAY from this data):
> [[System - Legacy Reporting]]
[Enter blank to finish]
Phase 5: Data Lineage (Optional)
Ask: "Track data lineage? (Y/n)"
If YES:
Derived From (upstream data sources):
> [[DataAsset - Source Invoices]]
[Enter blank to finish]
Feeds Into (downstream data assets):
> [[DataAsset - Revenue Dashboard]]
[Enter blank to finish]
Phase 6: Operational Details (Quick or Full)
Ask: "Quick capture or full detail? (Q/f)"
If Quick - skip to Phase 7 with defaults
If Full:
Refresh Frequency:
- real-time | hourly | daily | weekly | monthly | ad-hoc
Default: daily
Record Count (approximate): [number]
Volume per Day: [e.g., "2.5GB", "500K records"]
Retention Period: [e.g., "7 years", "90 days"]
Data Quality Metrics:
- Completeness (%): [e.g., 98.5]
- Uniqueness (%): [e.g., 99.9]
- Accuracy: high | medium | low
- Timeliness: [e.g., "< 5 minutes"]
SLAs:
- Availability: [e.g., "99.9%"]
- Latency: [e.g., "< 500ms"]
Governance:
- GDPR Applicable: Y/n
- PII Fields: [comma-separated list]
Phase 7: Generate Frontmatter
type: DataAsset
title: "{{name}}"
assetId: "{{assetId}}"
# Classification
domain: {{domain}}
dataType: {{dataType}}
classification: {{classification}}
# Location & Format
sourceSystem: "[[System - {{system}}]]"
storageLocation: "{{location}}"
format: {{format}}
# Ownership
owner: "[[{{owner}}]]"
steward: {{steward}}
# Relationships - Current State
producedBy: ["[[System - {{system}}]]"]
consumedBy: [{{consumers}}]
exposedVia: [{{exposedVia}}]
# Relationships - Future State
plannedConsumers: [{{plannedConsumers}}]
deprecatingConsumers: [{{deprecatingConsumers}}]
# Lineage
derivedFrom: [{{derivedFrom}}]
feedsInto: [{{feedsInto}}]
# Operational Metrics
refreshFrequency: {{frequency}}
recordCount: {{recordCount}}
volumePerDay: "{{volume}}"
retentionPeriod: "{{retention}}"
# Data Quality
completeness: {{completeness}}
uniqueness: {{uniqueness}}
accuracy: {{accuracy}}
timeliness: "{{timeliness}}"
# SLAs
slaAvailability: "{{slaAvailability}}"
slaLatency: "{{slaLatency}}"
# Governance
gdprApplicable: {{gdpr}}
piiFields: [{{piiFields}}]
# Quality Indicators
confidence: medium
freshness: current
verified: false
reviewed: null
created: {{today}}
modified: {{today}}
tags: [type/data-asset, domain/{{domain}}]
Phase 8: Generate Body Content
Create structured body with:
- Overview - What data this contains, business purpose
- Source Details - Table with system, location, format
- Data Relationships - Mermaid diagram showing producers/consumers
- Consumer Systems - Table of current, planned, and deprecating consumers
- Data Lineage - Upstream/downstream flow (if captured)
- Operational Metrics - Volume, refresh, quality (if full detail)
- Security & Governance - Classification, PII, GDPR status
- Related Notes - Links to ADRs, integrations, system notes
Phase 9: Create File
Filename: DataAsset - {{name}}.md
Location: Vault root
Output:
✅ Created: DataAsset - {{name}}.md
Linked to:
- [[System - {{producing system}}]] (producer)
- [[System - {{consumer1}}]] (consumer)
- [[System - {{consumer2}}]] (consumer)
Relationship summary:
- 1 producer
- {{n}} current consumers
- {{n}} planned consumers
- {{n}} deprecating consumers
Next steps:
1. Add to [[MOC - Data Assets]]
2. Create integration notes: /integration {{source}} {{target}}
3. Document data contract if critical
Example Interaction
User: /dataasset "Work Orders"
System: Which system produces this data?
> ERP System
Asset ID suggestion: ERP-WORK-ORDERS-001
> [accept]
Data Type:
> database-table
Domain:
> engineering
Classification:
> internal
Storage Location:
> erp.dbo.work_orders
Format:
> sql
Data Owner:
> [[Data Owner Name]]
Data Steward:
> [skip]
Current Consumers:
> [[System - Data Warehouse]]
> [[System - Analytics Platform]]
> [done]
Planned Consumers:
> [[System - New BI Tool]]
> [done]
Deprecating Consumers:
> [[System - Legacy Reporting]]
> [done]
Track lineage? (Y/n)
> n
Quick or Full detail? (Q/f)
> q
✅ Created: DataAsset - Work Orders.md
Linked to:
- [[System - ERP System]] (producer)
- [[System - Data Warehouse]] (consumer)
- [[System - Analytics Platform]] (consumer)
Relationship summary:
- 1 producer
- 2 current consumers
- 1 planned consumer
- 1 deprecating consumer
Key Differentiators from /datasource
| Aspect | /datasource | /dataasset |
|---|---|---|
| Future state | Not tracked | plannedConsumers, deprecatingConsumers |
| Lineage | Schema-level only | System-level derivedFrom/feedsInto |
| Detail level | Always full | Quick vs Full capture modes |
| Ownership | Single owner | owner + steward |
| SLAs | Not tracked | slaAvailability, slaLatency |
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
Didn't find tool you were looking for?