Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.anomalyarmor.ai/llms.txt

Use this file to discover all available pages before exploring further.

The /armor:lineage skill helps you understand how data flows through your systems by showing upstream sources and downstream consumers.

​
Usage

/armor:lineage
Or ask naturally:
  • “What depends on this table?”
  • “Where does this data come from?”
  • “Show me the impact of changing the customers table”

​
Concepts

​
Upstream

Tables that feed data into the target table (data sources).

​
Downstream

Tables that consume data from the target table (data consumers).

​
Depth

How many levels of dependencies to traverse:
  • depth=1: Direct dependencies only
  • depth=2: Two levels
  • depth=3+: Extended chain (max 5)

​
Example Usage

​
Find Downstream Consumers

You: What depends on the customers table?

Claude: Found 7 downstream tables that depend on customers:

Direct consumers (depth 1):
- gold.customer_metrics
- gold.order_summary
- reporting.customer_dashboard

Indirect consumers (depth 2):
- ml.churn_features
- ml.ltv_predictions
- analytics.weekly_cohorts
- external.bi_tool_view

Any changes to customers will affect these tables.

​
Find Upstream Sources

You: Where does the orders_mart table get its data?

Claude: orders_mart has 4 upstream sources:

Direct sources:
- staging.orders_raw
- staging.customers
- staging.products

Root sources:
- raw.events

Data flow:
raw.events -> staging.orders_raw -> gold.orders_mart

​
Impact Analysis

You: If I change the users table schema, what's affected?

Claude: Impact analysis for users:

Directly affected (12 tables):
- gold.user_metrics
- gold.user_activity
- reporting.user_dashboard
- ml.user_features
...

Indirectly affected (5 tables):
- analytics.monthly_actives
- reporting.exec_summary
...

Total: 17 tables may need updates.

Recommendation: Test changes in staging before production.

​
Common Questions

​
Where does lineage come from, query logs or dbt manifests?

AnomalyArmor parses query history from your warehouse to build table-level lineage automatically. If you use dbt, manifest-based lineage can be ingested on top for more complete column-level detail. See lineage.

​
How deep can I trace dependencies in one query?

Up to 5 levels. Ask for depth=3 or “show me 5 levels deep” to extend traversal. Deeper queries take longer to return, so start with depth 2 unless you’re doing full impact analysis.

Investigate

Root cause analysis

Status

Health overview