Govern and discover records with common metadata in AWS Agent Registry
AWS Agent Registry now supports custom metadata schemas, letting administrators define organization‑wide key‑value attributes (e.g., business unit, risk classification) that every MCP server, skill, or agent record must adhere to. Records are validated against the schema at creation or update, yielding a COMPLIANT or NON_COMPLIANT status, and the Discovery APIs can filter on these custom fields. This enables consistent governance, searchable catalogs, and automated discovery for large agentic fleets.
⚡ Key Takeaways
- Custom metadata schema fields can be of type Text, Enum, URL, or Boolean and may be marked required or optional.
- The registry enforces schema compliance, rejecting invalid submissions with a ValidationException.
- Records without custom metadata are stored as NON_COMPLIANT and cannot be approved until required fields are supplied.
- Discovery APIs accept filters on custom attributes, allowing agents and humans to locate resources by standardized facts.
- Compliance status (COMPLIANT vs NON_COMPLIANT) reflects schema adherence only; approval is a separate step.
Engineers can now programmatically enforce organizational standards across all MCP servers and skills, reducing manual governance overhead and improving the reliability of automated agent discovery pipelines.
✅ Practical Steps
- Define a custom metadata schema in AWS Agent Registry by specifying field names, types (Text/Enum/URL/Boolean), allowed values, and required flags.
- When publishing an MCP server record, include values for the defined schema fields; the registry will validate and return COMPLIANT status or a ValidationException.
- Use the Discovery APIs with filter parameters on custom attributes to retrieve compliant records for consumption by agents or UI tools.
Want the full story? Read the original article.
Read on AWS ML Blog ↗