metadata

Connection metadata introspection — record types, sObjects, tables, and field/column definitions fetched live from the remote system through a configured Celigo connection.

REST API: no dedicated spec — metadata hits adaptor-specific paths under /v1/netsuite/metadata/…, /v1/salesforce/metadata/…, and the generic PUT /v1/connections/{_id}/metadata RDBMS bridge. See the specs index and connection.yml for the underlying connection shape.

celigo metadata <subcommand> <connectionId> [entityType] [flags]

Supports all global flags. Auto-detects the adaptor from the connection's type — only netsuite, salesforce, and rdbms connections are supported.

---

Subcommands

Subcommand	Purpose
types <connectionId>	List entity types on a connection (NetSuite record types + saved searches, Salesforce sObject types, or RDBMS tables).
fields <connectionId> <entityType>	List fields/columns for a single entity (NetSuite record type, Salesforce sObject, or RDBMS table).

---

celigo metadata types <connectionId>

List the entity types available on a live connection. The CLI first GETs the connection, reads its type, and then dispatches to the correct adaptor endpoint.

Signature

celigo metadata types <connectionId> [--refresh]

Arguments

Argument	Type	Required	Description
<connectionId>	string	Yes	Connection ID. Must resolve to a connection whose type is netsuite, salesforce, or rdbms.

Flags

Flag	Type	Default	Description
--refresh	boolean	false	Bypass the server-side metadata cache and fetch fresh data from the remote system. Applied as ?refreshCache=true for NetSuite/Salesforce and as refreshCache: true in the PUT body for RDBMS.

Example

# NetSuite — record types + saved searches, merged into one list
celigo metadata types 5f83a9b2c7d3e8f1a2b3c4d5

# Salesforce — sObject types
celigo metadata types 5f83a9b2c7d3e8f1a2b3c4d5 --format json

# RDBMS — tables (force a fresh pull)
celigo metadata types 5f83a9b2c7d3e8f1a2b3c4d5 --refresh

Corresponds to (dispatched by connection type):

• NetSuite: GET /v1/netsuite/metadata/suitescript/connections/{connectionId}/recordTypes and GET /v1/netsuite/metadata/suitescript/connections/{connectionId}/savedSearches
• Salesforce: GET /v1/salesforce/metadata/connections/{connectionId}/sObjectTypes
• RDBMS: PUT /v1/connections/{connectionId}/metadata with body { "rdbms": { "type": "tables", "tables": "" }, "refreshCache": <bool> }

---

celigo metadata fields <connectionId> <entityType>

List fields (or columns) for a single entity. For Salesforce, the response includes related record types alongside the fields.

Signature

celigo metadata fields <connectionId> <entityType> [--refresh]

Arguments

Argument	Type	Required	Description
<connectionId>	string	Yes	Connection ID (netsuite, salesforce, or rdbms).
<entityType>	string	Yes	The entity to inspect. NetSuite: record-type id (e.g. customer, salesorder). Salesforce: sObject API name (e.g. Account, Contact). RDBMS: fully qualified table name (e.g. analytics.public.orders).

Flags

Flag	Type	Default	Description
--refresh	boolean	false	Bypass the cache and fetch fresh field metadata.

Example

# NetSuite record type
celigo metadata fields 5f83a9b2c7d3e8f1a2b3c4d5 customer

# Salesforce sObject
celigo metadata fields 5f83a9b2c7d3e8f1a2b3c4d5 Account --format json

# RDBMS (Snowflake) — must be fully qualified
celigo metadata fields 5f83a9b2c7d3e8f1a2b3c4d5 analytics.public.orders --refresh

Corresponds to (dispatched by connection type):

• NetSuite: GET /v1/netsuite/metadata/suitescript/connections/{connectionId}/recordTypes/{entityType}
• Salesforce: GET /v1/salesforce/metadata/connections/{connectionId}/sObjectTypes/{entityType}
• RDBMS: PUT /v1/connections/{connectionId}/metadata with body { "rdbms": { "type": "columns", "tables": "<entityType>" }, "refreshCache": <bool> }

---

Gotchas

• Only three connection types are supported. The CLI pre-flights with GET /v1/connections/{id} and rejects anything that isn't netsuite, salesforce, or rdbms with: Connection <id> has type "<type>" — metadata is only supported for netsuite, salesforce, rdbms connections.
• --refresh invalidates the cached metadata. Without it you get whatever the server cached on the last introspection. Pass it after a schema change in the remote system (new NetSuite custom field, new Salesforce custom object, new RDBMS column) to avoid stale pickers.
• NetSuite types merges two endpoints. Record types and saved searches come back as a single list with a synthesized _metaType column (recordType or savedSearch) so you can tell them apart.
• Salesforce fields returns fields and record types together. The response is an object with both — don't expect a flat array.
• RDBMS goes through the generic connection metadata bridge. NetSuite and Salesforce have dedicated GET endpoints; RDBMS is a PUT /v1/connections/{id}/metadata with a rdbms.type of tables or columns. The verb is PUT by design, not a typo.
• RDBMS tables must be fully qualified. For Snowflake/BigQuery/Postgres the entity name must be database.schema.table (or whatever the connection's default schema expects). If the call returns an empty column list, the CLI prints a yellow hint pointing at qualification and column-metadata permissions (e.g. BigQuery bigquery.tables.get on the dataset).
• The connection must be reachable. Metadata is fetched live — an offline or failing connection will return an error from the remote system, not an empty list. Run celigo connections ping <id> first when in doubt.

Related

• connections — the <connectionId> argument; also where you configure the credentials the metadata calls ride on.
• imports — field pickers in import mappings consume the same metadata.
• exports — export field selection (NetSuite record-type fields, Salesforce sObject columns, RDBMS table columns) mirrors what metadata fields returns.