From oracle-ai-data-platform-workbench-engineer-agent
Live browsing of AIDP data catalogs: list catalogs/schemas/tables/volumes, inspect columns, and resolve names to catalog keys. Ad-hoc, no SQL required.
How this skill is triggered — by the user, by Claude, or both
Slash command
/oracle-ai-data-platform-workbench-engineer-agent:aidp-catalog-exploreThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Browse and inspect the AIDP catalog directly via the control-plane — no SQL, no compute, no MCP required.
aidp-catalog-explore — live catalog browsingBrowse and inspect the AIDP catalog directly via the control-plane — no SQL, no compute, no MCP required.
Complements aidp-catalog-init: use this for ad-hoc, live lookups; use catalog-init to build the cached
grounding file.
aidp CLI (control-plane, read-only)Preferred engine is the official Oracle aidp CLI; oci raw-request is the fallback when the CLI isn't
installed. Both hit the same data-plane REST API with the same auth — see
references/aidp-cli-map.md for the full command map and
references/oci-raw-request.md for base URL + auth ladder.
| Goal | CLI (preferred) | REST fallback (verified) |
|---|---|---|
| List catalogs | aidp catalog list | GET /catalogs |
| Describe a catalog | aidp catalog get --catalog-key <cat> | (in list output) |
| List schemas in a catalog | aidp schema list --catalog-key <cat> | GET /schemas?catalogKey=<cat> |
| List tables in a schema | aidp schema list-tables --catalog-key <cat> --schema-key <cat.schema> | GET /tables?catalogKey=<cat>&schemaKey=<cat.schema> |
| Describe a single table | aidp schema get-table --catalog-key <cat> --schema-key <cat.schema> --table-key <t> | GET /tables?… then filter by table key client-side |
| List volumes | aidp volume list --catalog-key <cat> | GET /volumes?catalogKey=<cat> (param shape TBD — if 400, the error names the required param) |
All CLI calls take --instance-id <DATALAKE_OCID> --auth api_key --profile DEFAULT --region <r>.
# CLI (preferred): list tables in cat.schema
aidp schema list-tables --catalog-key default --schema-key default.default \
--instance-id <DATALAKE_OCID> --auth api_key --profile DEFAULT --region us-ashburn-1
# Fallback (no CLI installed): oci raw-request, --profile DEFAULT (api_key)
oci raw-request --http-method GET \
--target-uri "https://aidp.us-ashburn-1.oci.oraclecloud.com/20240831/dataLakes/<OCID>/tables?catalogKey=default&schemaKey=default.default" \
--profile DEFAULT
On 401/403/"Security Token" follow the auth ladder. Per-endpoint REST params are required — a bare
path returns 400 InvalidParameter: query param X must not be null, which names the missing param. REST
output is JSON: { "data": …, "headers": …, "status": <int> }.
aidp catalog list → aidp schema list --catalog-key <cat> →
aidp schema list-tables --catalog-key <cat> --schema-key <cat.schema> → aidp schema get-table (or the
REST fallback GET /catalogs → GET /schemas?catalogKey= → GET /tables?… → filter for the table)..aidp/catalog.md Quick Reference first (instant); fall back to scanning
/tables names + table objects for the concept..aidp/catalog.md for repeated lookups; this skill is for live/uncached inspection or
when the cache may be stale (then suggest aidp-catalog-init --refresh).aidp MCP is configured, its list_catalogs / list_schemas / list_tables / get_table /
list_volumes tools are an optional accelerator — not required.aidp CLI command map (primary engine)npx claudepluginhub anthropics/claude-plugins-official --plugin oracle-ai-data-platform-workbench-engineer-agentDiscovers and caches AIDP catalog metadata (tables, columns, joins) into a grounding file for NL-to-SQL. Run on first setup or after schema changes.
Full inventory and audit of AWS Glue Data Catalog assets across S3 Tables, Redshift-federated, and remote Iceberg catalogs. Use for catalog overview, listing tables, or data landscape mapping.
Searches DataHub catalog to discover datasets, find entities by platform/domain, and answer ad-hoc questions about metadata ownership and PII.