---
id: dwd_cb2bgvae
type: pattern
slug: active-metadata-consumption
title: "Serving metadata to people, systems and agents"
summary: "One canonical model, one governed write path, many read paths. How to turn catalog changes into subscribable domain events, and which channel serves which consumer."
status: draft
classification: public
authors:
  - name: Shan Umasankar
    url: https://dealwithdata.com
    role: author
assisted_by: [Claude]
scales: [enterprise]
tags: [architecture, events, ai, mcp, kafka]
terms: [dwd_mgglmnwq, dwd_xbqpgs7h, dwd_2agj6wti, dwd_qywin2zi, dwd_5k2ibrbv, dwd_tthoszn7, dwd_gdefyuzz]
related: [dwd_q3gbjug4, dwd_5hhpphgh, dwd_wg5zzqg2]
sources:
  - title: "OpenLineage"
    url: https://openlineage.io/
    accessed: 2026-10-07
  - title: "CloudEvents specification"
    url: https://cloudevents.io/
    accessed: 2026-10-07
  - title: "AsyncAPI"
    url: https://www.asyncapi.com/
    accessed: 2026-10-07
  - title: "Model Context Protocol"
    url: https://modelcontextprotocol.io/
    accessed: 2026-10-07
  - title: "Open Data Contract Standard (Bitol)"
    url: https://bitol-io.github.io/open-data-contract-standard/
    accessed: 2026-10-07
created: 2026-10-07
updated: 2026-10-07
version: 1
---

Once a catalog has good business metadata, everyone asks for it: "Can we
get technical metadata with business metadata?", "Can our AI agent use the
definitions?" The failure mode is ten teams building ten extracts. The fix
is a shape, not a tool.

## The shape

**One canonical model, one governed write path, many read paths.**

The hard, valuable part isn't the streaming. It's the governed link:
[[business-term]] → [[business-element]] → physical column → lineage.
Everything below is delivery.

## Getting changes out

Most catalogs sit on a relational database and are poll or harvest based.
Changes arrive two ways:

- **UI edits** (a steward changes a definition): capture with database
  [[change-data-capture]] such as GoldenGate or Debezium.
- **Harvests** (scanners reload technical metadata): diff successive
  snapshots and emit a change set. This gives cleaner events than the
  database log.

## Three layers, not one

```
Catalog DB ──CDC──▶ raw.* topics          (private: one team reads these)
Harvests ──diff──▶        │
                          ▼
                    Translator            row changes → domain events
                          │               schema registry, versioned
                          ▼
              public topics               BusinessElementDefinitionChanged
                                          LineageEdgeAdded
                                          CdeStewardReassigned
```

Never expose raw CDC from a vendor's internal schema. It is undocumented,
changes on upgrade, and one business edit fires a dozen row changes.
Subscribers break and drown.

## Pick the channel by consumer

| Consumer | Channel | Standard |
|---|---|---|
| Stewards, analysts | Catalog UI | n/a |
| Applications | REST or GraphQL over the metadata graph | OpenAPI |
| Change subscribers | Domain-event topics | CloudEvents envelope, AsyncAPI docs |
| Bulk analytics | Metadata snapshots as tables | SQL |
| AI agents | MCP server plus a search index | MCP |
| Pipelines | Emit lineage at run time; enforce contracts in CI | OpenLineage, ODCS |
| BI and AI tools | Semantic model export | OSI (Apache Ossie) |

All of these are projections of the same model. When a team asks for "a
feed", they get one of these, not a custom extract.

## Rules that keep it trustworthy

1. **Stable global IDs** for every term, element and physical asset, or the
   links can't survive across systems.
2. **The link is a governed object.** A business-to-technical mapping carries
   an owner, a confidence score, provenance (human, rule or AI) and a status.
   AI-proposed mappings enter as *proposed*; stewards authorize.
3. **The catalog stays the system of record for its domain.** Publish
   outward; don't try to make it the hub for everything.
4. **Agents read through the MCP layer, never the database.** That gives one
   place for entitlements, audit, and "approved definitions only".

## When it becomes a nervous system

[[active-metadata]] only matters if something depends on it. Wire in two or
three consumers that break when metadata is wrong, such as change
management, a regulatory report pipeline, and one AI agent. Until then,
it's documentation.

## Using LLMs for long descriptions

For bulk drafting of element descriptions, a fast, cheap model is the right
default. Quality comes from grounding, not model size: feed the physical
name and type, profile statistics, lineage neighbors, the parent dataset
and nearby glossary terms. Route critical elements and low-confidence drafts
to a stronger model or a human reviewer.