fabio-lakehouse

name	fabio-lakehouse
description	Intent-scoped fabio skill for Microsoft Fabric Lakehouse work: create lakehouses, upload/load/sync files and Delta tables, OneLake shortcuts, Iceberg, and Materialized Lake Views. Use when the user wants to ingest, organize, or move data in a lakehouse. Triggers: "create lakehouse", "upload to lakehouse", "load table", "list tables", "sync lakehouse", "onelake", "shortcut", "materialized lake view".
license	MIT

fabio-lakehouse — Lakehouse — files, tables, sync, Iceberg, Materialized Lake Views

Generated file — do not edit by hand. This intent-scoped sub-skill of the fabio skill is generated from fabio's command schema plus authored judgment. Regenerate with cargo test generate_subskills -- --ignored. For install, auth, output envelope, global flags, and agent-safety rules, see the root fabio skill.

Prefer runtime introspection. This index is a snapshot; the installed binary is always authoritative. Use fabio context agent --group <group> and fabio context describe <group> <command> for exact flags and output shapes.

When to use

• Creating a lakehouse or listing/inspecting its files and Delta tables.
• Uploading files to Files/ and loading them into Delta tables.
• Syncing files between lakehouses or from a local directory (rsync-like).
• Creating OneLake shortcuts (typed flags for all 9 targets: OneLake, ADLS Gen2, S3, GCS, S3-compatible, Dataverse, Azure Blob, External Data Share, OneDrive/SharePoint), listing them (list-shortcuts), or managing OneLake security.
• Deleting an arbitrary OneLake directory recursively (delete-directory), beyond just delete-table.
• Managing Materialized Lake Views (execution definitions + refresh schedules).

When NOT to use (route elsewhere)

• Querying lakehouse tables with T-SQL -> use fabio-warehouse-sql (sql-endpoint / lakehouse plan/query).
• Real-time/streaming ingestion or KQL -> use fabio-rti-kql.
• Report/semantic model over the lakehouse -> use the bi-developer persona.

Command index

Generated from fabio's command schema. For full flag details use fabio context agent --group <group> or fabio context describe <group> <command>.

fabio lakehouse

Manage lakehouses (tables, files, shortcuts)

Command	Mutates	Description
fabio lakehouse bulk-create-shortcuts	yes	Bulk-create multiple shortcuts (LRO)
fabio lakehouse copy-file	yes	Copy files between lakehouses (supports glob patterns for parallel copy)
fabio lakehouse copy-table	yes	Copy a table between lakehouses
fabio lakehouse create	yes	Create a new lakehouse
fabio lakehouse create-directory	yes	Create a directory in a lakehouse (DFS)
fabio lakehouse create-execution-definition	yes	Create a materialized lake view execution definition
fabio lakehouse create-materialized-views-schedule	yes	Create a schedule for materialized lake view refresh
fabio lakehouse create-shortcut	yes	Create a shortcut (typed target flags per target type, or raw --target JSON)
fabio lakehouse delete	yes	Delete a lakehouse
fabio lakehouse delete-directory	yes	Recursively delete a directory from a lakehouse (DFS; irreversible)
fabio lakehouse delete-execution-definition	yes	Delete a materialized lake view execution definition
fabio lakehouse delete-file	yes	Delete a file from a lakehouse
fabio lakehouse delete-materialized-views-schedule	yes	Delete a schedule for materialized lake view refresh
fabio lakehouse delete-shortcut	yes	Delete a shortcut
fabio lakehouse delete-table	yes	Delete a table from a lakehouse
fabio lakehouse download	no	Download a file from a lakehouse
fabio lakehouse get-definition	no	Get the definition of a lakehouse
fabio lakehouse get-livy-session	no	Get details of a Livy session for a lakehouse
fabio lakehouse get-shortcut	no	Get shortcut details
fabio lakehouse iceberg-config	no	Get Iceberg REST Catalog configuration for a lakehouse
fabio lakehouse iceberg-credentials	no	Load vended storage credentials scoped to a specific table
fabio lakehouse iceberg-namespace	no	Get metadata for a specific namespace via the Iceberg REST Catalog
fabio lakehouse iceberg-namespace-exists	no	Check if a namespace exists via the Iceberg REST Catalog (lightweight HEAD)
fabio lakehouse iceberg-namespaces	no	List table namespaces (schemas) via the Iceberg REST Catalog
fabio lakehouse iceberg-snapshots	no	Show snapshot history for a table via the Iceberg REST Catalog
fabio lakehouse iceberg-stats	no	Show table statistics from the latest Iceberg snapshot (record/file counts, size)
fabio lakehouse iceberg-table	no	Get table definition (schema, partitions, properties) via the Iceberg REST Catalog
fabio lakehouse iceberg-table-exists	no	Check if a table exists via the Iceberg REST Catalog (lightweight HEAD)
fabio lakehouse iceberg-tables	no	List tables in a namespace via the Iceberg REST Catalog
fabio lakehouse list	no	List lakehouses in a workspace
fabio lakehouse list-execution-definitions	no	List materialized lake view execution definitions for a lakehouse
fabio lakehouse list-files	no	List files in a lakehouse
fabio lakehouse list-livy-sessions	no	List Livy sessions for a lakehouse
fabio lakehouse list-shortcuts	no	List shortcuts in an item (hides DW-managed shortcuts by default)
fabio lakehouse list-tables	no	List tables in a lakehouse
fabio lakehouse load-table	yes	Load a file (already in the lakehouse) into a Delta table
fabio lakehouse move-file	yes	Move files between lakehouses (supports glob patterns for parallel move)
fabio lakehouse move-table	yes	Move a table between lakehouses (copy + delete source)
fabio lakehouse optimize-table	yes	Optimize a Delta table (V-Order compaction + optional Z-Order)
fabio lakehouse plan	yes	Capture the estimated execution plan (SHOWPLAN_XML) without executing the query
fabio lakehouse queries-frequent	yes	List frequently-run queries (from queryinsights.frequently_run_queries)
fabio lakehouse queries-history	yes	List completed query history (from queryinsights.exec_requests_history)
fabio lakehouse queries-long-running	yes	List long-running queries (from queryinsights.long_running_queries)
fabio lakehouse queries-running	yes	List currently running queries on the lakehouse SQL endpoint
fabio lakehouse query	no	Execute SQL against the lakehouse SQL endpoint
fabio lakehouse refresh-materialized-views	yes	Trigger a refresh of materialized lake views
fabio lakehouse run-table-maintenance	yes	Run table maintenance on a lakehouse
fabio lakehouse show	no	Show details of a lakehouse
fabio lakehouse show-execution-definition	no	Show a materialized lake view execution definition
fabio lakehouse sync	yes	Sync files between lakehouses (parallel, copies new/modified files)
fabio lakehouse table-schema	no	Show Delta table schema (reads from OneLake _delta_log without Spark/SQL)
fabio lakehouse update	yes	Update a lakehouse (rename/redescribe)
fabio lakehouse update-definition	yes	Update the definition of a lakehouse
fabio lakehouse update-execution-definition	yes	Update a materialized lake view execution definition
fabio lakehouse update-materialized-views-schedule	yes	Update a schedule for materialized lake view refresh
fabio lakehouse upload	yes	Upload files to a lakehouse (supports glob patterns for parallel upload)
fabio lakehouse upload-table	yes	Upload a local file and load it into a Delta table (upload + load-table in one step)
fabio lakehouse vacuum-table	yes	Vacuum a Delta table (remove old files beyond retention period)

fabio onelake-security

Manage OneLake data access roles (row/column-level security)

Command	Mutates	Description
fabio onelake-security create	yes	Create or update a single data access role
fabio onelake-security delete	yes	Delete a data access role
fabio onelake-security list	no	List data access roles for an item
fabio onelake-security show	no	Show details of a data access role
fabio onelake-security upsert	yes	Replace all data access roles for an item (atomic PUT)

Must / Prefer / Avoid

MUST

• Use PascalCase for --mode (Overwrite/Append) and --format (Csv/Parquet).
• Upload to Files/ before load-table, or use upload-table for the combined one-step path.
• Convert JSON to CSV/Parquet before load-table (JSON is not a supported load format).

PREFER

• lakehouse sync (ETag-based, rename-aware) over manual re-upload loops.
• move-file within a lakehouse (atomic O(1) rename) over copy+delete.
• Runtime introspection (context agent --group lakehouse, context describe) over guessing flags.

AVOID

• load-table with JSON format (unsupported).
• Lowercase enum values like --mode overwrite or --format csv (silently rejected by the API).
• Deleting Lakehouse/Warehouse/other data-bearing items without explicit user approval.

Key gotchas

• Table file listing lists from the root to get real item-id-prefixed paths.
• move-file within a lakehouse is an atomic O(1) rename; cross-item falls back to copy+delete automatically.
• list-shortcuts hides DW-managed internal shortcuts (OneLake->OneLake under Tables/) by default — pass --include-managed to see them.
• create-shortcut takes typed flags per target type (--connection-id/--location/--subpath for cloud; --target-workspace/--target-item/--target-path for OneLake; --bucket for S3); --target JSON remains a raw escape hatch.

Troubleshooting

Symptom	Fix
load-table fails or ignores the file	Check --mode/--format are PascalCase (Overwrite, Csv) and the format is Csv or Parquet, not JSON.
FORBIDDEN when creating/deleting	You need at least Member role on the workspace; Delete requires Member+.
list-tables shows no rows after an ETL run	Confirm the notebook/job actually wrote Delta tables and completed (use notebook run --wait).
move-file returns 403 across items	Atomic rename only works within the same lakehouse; across items fabio falls back to copy+delete automatically.

Safety

• load-table --mode Overwrite replaces the whole table — confirm with the user.
• sync --delete removes destination files not present in source — preview without --delete first.
• delete-directory is a recursive OneLake delete — fabio refuses an empty/root path (that would erase the entire lakehouse); confirm the exact subpath, and it is --dry-run-guarded.
• Deleting a lakehouse is a protected, data-bearing operation — never do it without explicit user approval.

Shared references

Cross-cutting operational guidance (the "common" layer) — consult the relevant topic before non-trivial work:

Reference	Covers
fabio context best-practices throttling	fabio transparently handles 429 (Too Many Requests) and gateway errors. Agents do NOT need to implement retry logic.
fabio context best-practices pagination	fabio handles pagination via --all (auto-fetch all pages), --continuation-token (resume), and --limit (truncate). Agents rarely need to paginate manually.
fabio context best-practices lro	Many Fabric operations are async (return 202). fabio polls them automatically. Use --wait for job operations.
fabio context best-practices shortcuts	OneLake shortcuts virtualize external/other-OneLake storage into a lakehouse. Create with TYPED flags per target type (9 supported: OneLake, AdlsGen2, AmazonS3, AzureBlobStorage, GoogleCloudStorage, S3Compatible, Dataverse, ExternalDataShare, OneDriveSharePoint); cloud targets need a connection first. list-shortcuts enumerates them (DW-managed internal shortcuts hidden unless --include-managed). load-table resolves shortcut paths even when list-files does not show their contents.
fabio context best-practices lakehouse-mlv	Materialized Lake Views are pre-computed Delta table snapshots inside a Fabric Lakehouse that accelerate repeated query patterns. They are refreshed on demand or on a schedule, and their refresh behavior is controlled by MLV execution definitions.

See also

• fabio context persona data-engineer
• fabio context workflow lakehouse-etl
• fabio context workflow lakehouse-mlv
• fabio context disambiguate materialized-view
• fabio context disambiguate sql-endpoint