Deep-dives into Dynamics 365 entity metadata: fields with types, option sets, relationships, keys. Triggers on entity structure queries like 'show fields on X'.
How this skill is triggered — by the user, by Claude, or both
Slash command
/dynamics365-dataverse:explore-metadataThis skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
The user wants to explore or understand the metadata structure of their Dynamics 365 environment.
The user wants to explore or understand the metadata structure of their Dynamics 365 environment.
Argument provided: $ARGUMENTS
Call list_environments and ask the user which environment to explore. Call select_environment.
list_entities (optionally with custom_only: true) to show what's availablelist_entities with a filterFor the target entity, gather all metadata in parallel where possible:
get_entity_details → basic info (display name, logical name, primary ID/name, system vs custom)get_entity_attributes (with include_option_sets: true) → all fields with their types and option set valuesget_entity_relationships → all One-to-Many, Many-to-One, Many-to-Many relationshipsget_entity_keys → alternate keysFormat the output clearly — think of it like the tabs in the metadata viewer app:
Present as a table:
| Logical Name | Display Name | Type | Required | Custom |
|---|---|---|---|---|
| Sort by logical name. Highlight: |
For every Picklist / Status / State field, show the values:
| Value | Label |
|---|---|
| Group by field name. |
Show separately:
If any exist, list the key name and which fields compose it.
After presenting, ask if the user wants to:
new_, contoso_)EntitySetName is the plural name used in API URLs — this is what query_records needs_value suffix fields — these are the raw GUID values of lookup fieldsstatecode (State) and statuscode (Status Reason) exist on every entity — always show their option valuesnpx claudepluginhub nickmeron/dataverse-mcp-serverGuides creation, editing, and verification of skills for AI coding agents using test-driven development with subagent scenarios. Use when authoring or debugging skills.