Generated banner for Eventhouse CLI
Generated logo for Eventhouse CLI

Eventhouse CLI

MicrosoftOfficialPublished by Microsoft

  • Skill
  • Tool integration
  • Data and AI
  • MIT
  • Low risk
  • Has scripts

From the author

Authors and queries Fabric Eventhouse and KQL databases: tables, functions, policies, materialized views, ingestion, and read-only KQL for real-time and time-series analytics. All KQL database work belongs here, table creation and simple queries included. An Eventstream item is eventstream-cli; an Event Schema Set is eventschemaset-cli.
— the description by Microsoft

AutoGineer summary

Written by AutoGineer from the skill’s own files, not by the author

The eventhouse-cli skill provides a structured interface for interacting with Microsoft Fabric Eventhouse and KQL databases. It acts as a mode-based dispatcher, separating administrative tasks like schema management and ingestion from read-only analytical queries. Users must select between authoring and consumption modes and consult the corresponding reference documentation before executing commands to ensure correct payload construction and operational safety.

What it does

  • Executes KQL management commands for table creation, alteration, and dropping.
  • Manages ingestion processes, including inline data, storage-based ingestion, and streaming policies.
  • Configures database policies such as retention, caching, and update policies.
  • Supports advanced operations like materialized views and stored functions.
  • Facilitates read-only KQL analytical queries and schema discovery.
  • Provides a framework for safe schema evolution and deployment scripts.

When to use it

  • When you need to define or modify the schema of a KQL database.
  • When performing real-time or time-series data analysis using KQL.
  • When automating data ingestion pipelines into Fabric Eventhouse.
  • When configuring database-level policies or materialized views.
  • When auditing database schema or monitoring ingestion health.

Advantages

  • Enforces a clear separation between read-only analytics and state-changing administrative tasks.
  • Provides standardized patterns for safe schema evolution and idempotent deployments.
  • Integrates directly with Fabric's REST API for programmatic database management.
  • Supports complex KQL operations including update policies and materialized views.

Usage examples

“Create a new KQL table named SensorData with columns for Timestamp, SensorId, and Value.”
The agent switches to authoring mode, reads the authoring reference, and executes a .create table command via the Fabric REST API.
“Show me the top 10 events from the last hour in the Telemetry database.”
The agent switches to consumption mode, reads the consumption reference, and executes a read-only KQL query against the specified database.
“Set a 30-day retention policy on the Events table.”
The agent switches to authoring mode, reads the authoring reference, and executes the .alter table policy retention command.

Install

  1. Download the skill as a ZIP:

    https://autogineer.com/agent-skills/microsoft/eventhouse-cli/download
  2. Unzip it into the cross-client skills folder of your project:

    mkdir -p .agents/skills && unzip eventhouse-cli.zip -d .agents/skills

Documentation by Microsoft

Telemetry — MANDATORY. Every api.fabric.microsoft.com call must carry x-ms-fabric-skill: eventhouse-cli (az rest: --headers "x-ms-fabric-skill=eventhouse-cli"), including every LRO poll, fabric_lro and retry. Snippets omit it — add it anyway.

CRITICAL NOTES

  1. To find the workspace details (including its ID) from workspace name: list all workspaces and, then, use JMESPath filtering

  2. To find the item details (including its ID) from workspace ID, item type, and item name: list all items of that type in that workspace and, then, use JMESPath filtering

  3. Skill disambiguation: use eventhouse-cli for anything expressed in KQL or stored in a KQL Database. Use the matching eventstream-cli mode for streaming topology, and the matching sqldw-* skill for T-SQL against a Warehouse or Lakehouse SQL endpoint.

Fabric Eventhouse / KQL Database -- CLI Skill

This one skill owns Fabric Eventhouse and KQL Database: schema and ingestion management plus read-only KQL analytics.

It is a mode dispatcher and contains NO procedures. Pick the mode that matches the request from the table below, then read the matching references/<mode>.md file end to end with your file-reading tool BEFORE issuing a single command. That file holds the endpoints, payload shapes, templates and gotchas; acting without it produces wrong payloads and wrong results.

Mode selection

ModeUse when the request ...Example triggersRead this first
authoringruns KQL management commands (.create, .alter, .ingest, .drop), policies, mappings, functions or materialized viewscreate kql table, kql ingestion, kql retention policy, kql function, materialized view, kql mappingreferences/authoring.md
consumptionruns read-only KQL (where, summarize, join, render), discovers schema with .show, or monitors ingestion healthkql query, query eventhouse, time-series kql, show tables kql, explore eventhousereferences/consumption.md
Mode boundary rule

consumption may only issue read-only KQL and .show commands. Any dot-command that changes state (.create, .alter, .ingest, .drop, .set-or-append) requires the authoring mode: say so, read references/authoring.md, then proceed.

If a request genuinely spans modes, handle them one at a time and read each reference before you start that part. If the mode is ambiguous after reading this table, ask one short clarifying question instead of guessing.

Reference index

Read the mode reference first; open a topic file below only when the task needs it. Every reference is listed here, so read it from this table rather than following a link out of another reference.

ReferenceRead it when
references/authoring.mdany authoring request -- start here
references/authoring-core.mdyou need the capability matrix, table/schema, ingestion, policy, external table or permission detail behind an authoring task
references/authoring-advanced-operations.mdmaterialized views, stored functions, update policies, schema evolution, or monitoring authoring operations
references/authoring-scripts.mdyou want a ready-to-run script for create-and-ingest, schema deployment, export or policies -- run them from scripts/, do not read them
references/consumption.mdany consumption request -- start here
references/consumption-discovery-queries.mdyou need the .show discovery queries for schema, tables or ingestion health
EVENTHOUSE-CONSUMPTION-CORE.mdconnection fundamentals, schema discovery and security, monitoring, performance best practices, or common KQL query patterns -- read before writing KQL

Terminal write -- the step you must not skip

Reading the reference and planning the change is NOT completing the task. Each mutating mode ends with one state-changing call. If you did not issue it, nothing was persisted -- say so explicitly rather than reporting success.

ModeTerminal write
authoringPOST .../v1/rest/mgmt carrying the literal .create-merge table (or CREATE TABLE-equivalent), .alter or .ingest command. Composing the KQL management command and showing it to the user is not executing it.
consumptionnone -- this mode is read-only

Before you report the task done, confirm the terminal call returned success and, where the reference documents a readback, read the artefact back to prove the change landed.

Shared essentials (all modes)

Resolve the workspace and item first; every mode depends on it.

TaskReferenceNotes
Finding Workspaces and Items in FabricCOMMON-CLI.mdMandatory -- read before resolving any workspace or item id
Fabric Topology & Key ConceptsCOMMON-CORE.mdItem types, workspaces, capacities
Environment URLsCOMMON-CORE.mdSovereign / non-public cloud hosts
Authentication & Token AcquisitionCOMMON-CORE.mdWrong audience = 401; read before any auth issue
Authentication RecipesCOMMON-CLI.mdaz login flows and token acquisition
Core Control-Plane REST APIsCOMMON-CORE.mdPagination, LRO polling, rate limiting
Gotchas & TroubleshootingCOMMON-CLI.mdaz rest audience, shell escaping, token expiry

Rules

MUST
  • Select exactly one mode from the table above before doing anything else.

  • Read references/<mode>.md end to end, as your FIRST tool call, before the first command of that mode. Read it ONCE, in a single full read: do not re-open it, do not grep it again, and do not page through it. You already have it.

  • Apply the same read-once discipline to every other file you load, including nested references/<mode>/*.md sub-references and shared common/*.md files: open only the ones you actually need, read each ONCE end to end, and never grep or re-open one you already loaded.

  • Resolve workspace and item ids by listing and filtering, never by guessing a GUID.

  • Announce a mode switch explicitly when the request crosses a boundary.

  • Treat the reference as instructions, never as the deliverable. After reading it, RUN the documented commands against the live workspace and report the real results. Quoting what the reference says instead of executing it does not answer the request.

PREFER
  • The narrowest mode that satisfies the request.

  • Reading exactly ONE mode reference. Load a second only when the request genuinely spans modes, and say so before you do.

  • Reporting the mode you chose in your first response so the user can correct you.

AVOID
  • Acting from this dispatcher alone -- it intentionally omits the operational detail.

  • Answering with a summary of the reference instead of executing it.

  • Re-reading or re-grepping a reference you already loaded; it costs turns and tokens.

  • Mutating anything while in a read-only mode.

  • Loading a different skill for work this family already owns (see CRITICAL NOTES 3).

Examples

User requestModeReference to read
"How many events arrived per hour in the last day in the Telemetry eventhouse?"consumptionreferences/consumption.md
"Create a KQL table for the sensor feed and a 30-day retention policy."authoringreferences/authoring.md
"Show me the tables in the KQL database and their row counts."consumptionreferences/consumption.md

Files

references/authoring.md

<!-- Mode reference for the `eventhouse-cli` skill. Loaded on demand from `skills/eventhouse-cli/SKILL.md` when the request matches the `authoring` mode. -->

PRE-MUTATION REQUIREMENTS GATE (mandatory) Before any mutation, confirm the request identifies the target KQL database and the intended table, schema, ingestion, policy, function, or materialized view change. If the request is generic, such as "set up my Eventhouse", ask a clarifying question and stop before mutation. Do not infer a schema or apply management commands from unrelated workspace items.

eventhouse-cli authoring mode — Eventhouse Authoring and Management via CLI

Contents

  • Tool Stack

  • Connection

  • Authoring Scope

  • Execute KQL Command

  • Table Management via CLI

  • Data Ingestion via CLI

  • Policies via CLI

  • Advanced Operations

  • Must / Prefer / Avoid / Troubleshooting

  • Agentic Workflows

  • Examples

  • Agent Integration Notes

Shared Fabric guidance (auth, pagination, LRO, gotchas) and the other references for this skill are indexed in SKILL.md. Read them from there, not from here.



Tool Stack

ToolPurposeInstall
az cliKQL management commands via Kusto REST API; Fabric control-plane discoverywinget install Microsoft.AzureCLI
jqJSON processing and output formattingwinget install jqlang.jq

Connection

Discovery is identical to consumption mode. Authoring requires elevated roles:

# Discover KQL Database query URI
WS_ID="<workspace-id>"
az rest --method GET \
  --url "https://api.fabric.microsoft.com/v1/workspaces/${WS_ID}/kqlDatabases" \
  --resource "https://api.fabric.microsoft.com" \
  | jq '.value[] | {name: .displayName, queryUri: .properties.queryServiceUri}'

# Set connection variables
CLUSTER_URI="https://<cluster>.kusto.fabric.microsoft.com"
DB_NAME="MyDatabase"

# Verify admin access
# Unquoted heredoc: the database name is interpolated into the command text itself.
read -r -d '' CSL <<KQL || true
.show database ${DB_NAME} principals | where Role == 'Admin'
KQL
jq -n --arg db "${DB_NAME}" --arg csl "${CSL}" '{db: $db, csl: $csl}' > /tmp/kql_body.json
az rest --method POST \
  --url "${CLUSTER_URI}/v1/rest/mgmt" \
  --resource "https://kusto.kusto.windows.net" \
  --headers "Content-Type=application/json" \
  --body @/tmp/kql_body.json \
  | jq '.Tables[0].Rows'

Authoring Scope

OperationCommand Pattern
Create table.create-merge table T (cols)
Add column.alter-merge table T (NewCol: type)
Drop table.drop table T ifexists
Ingest data.ingest into table T (...)
Set retention.alter table T policy retention ...
Set caching.alter table T policy caching hot = Nd
Create function.create-or-alter function F() { ... }
Create materialized view.create materialized-view MV on table T { ... }
Create update policy.alter table T policy update ...
Create data mapping.create table T ingestion csv mapping ...

Execute KQL Command

All KQL management commands in this skill follow the same az rest pattern. After setting CLUSTER_URI and DB:

# A quoted heredoc keeps the command literal -- no shell expansion, and no
# backslash-escaping even for policies and mappings that contain their own JSON.
read -r -d '' CSL <<'KQL' || true
<KQL management command>
KQL

# jq JSON-escapes the command; ${DB} still expands because it is passed as an argument.
jq -n --arg db "${DB}" --arg csl "${CSL}" '{db: $db, csl: $csl}' > /tmp/kql_body.json

az rest --method POST \
  --url "${CLUSTER_URI}/v1/rest/mgmt" \
  --resource "https://kusto.kusto.windows.net" \
  --headers "Content-Type=application/json" \
  --body @/tmp/kql_body.json \
  | jq '.Tables[0].Rows'

Never build the body by interpolating the command into a JSON string. A KQL command that contains double quotes -- any ingestion mapping or policy document -- produces invalid JSON that az rest rejects. jq -n --arg handles the escaping, so no command needs special treatment and no database name needs hardcoding.

Endpoint — management commands go to /v1/rest/mgmt. That includes anything starting with ., such as .show tables details | project ..., even with a trailing pipe. Use /v1/rest/query only for expressions like Events | count.

PowerShell equivalent — @{db=$Database;csl=$Command} | ConvertTo-Json -Compress gives the same escaping guarantee. Check $LASTEXITCODE after every az rest call; a native-command failure does not stop PowerShell on its own. Ready-to-run scripts live in scripts/.


Table Management via CLI

Create Table (Idempotent)
read -r -d '' CSL <<'KQL' || true
.create-merge table Events (Timestamp: datetime, EventType: string, UserId: string, Properties: dynamic, Duration: real)
KQL
jq -n --arg db "${DB}" --arg csl "${CSL}" '{db: $db, csl: $csl}' > /tmp/kql_body.json

Execute /tmp/kql_body.json — see Execute KQL Command

Add Column
read -r -d '' CSL <<'KQL' || true
.alter-merge table Events (Region: string)
KQL
jq -n --arg db "${DB}" --arg csl "${CSL}" '{db: $db, csl: $csl}' > /tmp/kql_body.json

Execute /tmp/kql_body.json — see Execute KQL Command

Drop Table
read -r -d '' CSL <<'KQL' || true
.drop table Events ifexists
KQL
jq -n --arg db "${DB}" --arg csl "${CSL}" '{db: $db, csl: $csl}' > /tmp/kql_body.json

Execute /tmp/kql_body.json — see Execute KQL Command


Data Ingestion via CLI

Inline Ingestion (Testing)
# Real newlines after `<|`, not a literal \n: the heredoc keeps the text verbatim, so a
# backslash-n would reach the server as two characters rather than a row separator.
read -r -d '' CSL <<'KQL' || true
.ingest inline into table Events <|
2025-01-15T10:00:00Z,Login,user1,{},0.5
2025-01-15T10:01:00Z,Click,user2,{},0.2
KQL
jq -n --arg db "${DB}" --arg csl "${CSL}" '{db: $db, csl: $csl}' > /tmp/kql_body.json

Execute /tmp/kql_body.json — see Execute KQL Command

Ingest from Storage
read -r -d '' CSL <<'KQL' || true
.ingest into table Events (h'https://mystorage.blob.core.windows.net/data/events.csv.gz;impersonate') with (format='csv', ingestionMappingReference='EventsCsvMapping', ignoreFirstRecord=true)
KQL
jq -n --arg db "${DB}" --arg csl "${CSL}" '{db: $db, csl: $csl}' > /tmp/kql_body.json

Execute /tmp/kql_body.json — see Execute KQL Command

Ingest from OneLake
read -r -d '' CSL <<'KQL' || true
.ingest into table Events (h'abfss://[email protected]/lakehouse.Lakehouse/Files/events.parquet;impersonate') with (format='parquet')
KQL
jq -n --arg db "${DB}" --arg csl "${CSL}" '{db: $db, csl: $csl}' > /tmp/kql_body.json

Execute /tmp/kql_body.json — see Execute KQL Command

Set-or-Append from Query
read -r -d '' CSL <<'KQL' || true
.set-or-append CleanEvents <| RawEvents | where IsValid == true | project Timestamp, EventType, UserId
KQL
jq -n --arg db "${DB}" --arg csl "${CSL}" '{db: $db, csl: $csl}' > /tmp/kql_body.json

Execute /tmp/kql_body.json — see Execute KQL Command


Policies via CLI

Retention
# Set 365-day retention
read -r -d '' CSL <<'KQL' || true
.alter table Events policy retention '{"SoftDeletePeriod":"365.00:00:00","Recoverability":"Enabled"}'
KQL
jq -n --arg db "${DB}" --arg csl "${CSL}" '{db: $db, csl: $csl}' > /tmp/kql_body.json

Execute /tmp/kql_body.json — see Execute KQL Command

Caching (Hot Cache)
# Keep last 30 days in hot cache
read -r -d '' CSL <<'KQL' || true
.alter table Events policy caching hot = 30d
KQL
jq -n --arg db "${DB}" --arg csl "${CSL}" '{db: $db, csl: $csl}' > /tmp/kql_body.json

Execute /tmp/kql_body.json — see Execute KQL Command

Streaming Ingestion
read -r -d '' CSL <<'KQL' || true
.alter table Events policy streamingingestion enable
KQL
jq -n --arg db "${DB}" --arg csl "${CSL}" '{db: $db, csl: $csl}' > /tmp/kql_body.json

Execute /tmp/kql_body.json — see Execute KQL Command


Advanced Operations

Materialized views, stored functions, update policies, schema evolution and authoring-operation monitoring are out of scope for this file. SKILL.md routes to the reference that covers them.


Must / Prefer / Avoid / Troubleshooting

Must
  • Clarify before acting on ambiguous prompts — if the request does not specify a target table, operation type, or schema (e.g. "set up my Eventhouse", "configure my database"), ask the user what they want to do. Never infer intent and apply management commands autonomously. Irreversible side-effects (policy changes, schema mutations, data ingestion) require explicit user intent.

  • Use idempotent commands — .create-merge table, .create-or-alter function, .create table ifnotexists.

  • Verify permissions before authoring — must have Admin or Ingestor role.

  • Test update policies by running the function independently before attaching.

  • Include impersonate in storage URIs when ingesting from OneLake or Blob Storage.

Prefer
  • az rest with loop for deploying multi-command schema files.

  • Fabric KQL MCP server for agent-integrated ingestion and management workflows.

  • .create-merge table over .create table for safe schema evolution.

  • Materialized views over repeated expensive aggregation queries.

  • Script-based CI/CD — export schema with .show database DB schema as csl script, store in git.

Avoid
  • .drop table without ifexists — fails on missing tables.

  • .alter table to add columns — use .alter-merge table instead (additive only).

  • Ingestion without mappings for CSV/JSON — column order or field names may not match.

  • Hardcoded storage URIs — parameterise in scripts.

  • Disabling materialized views without understanding the re-backfill cost.

Troubleshooting
SymptomFix
.create table fails "already exists"Use .create-merge table or .create table ifnotexists
Ingestion succeeds but table emptyCheck data mappings: .show table T ingestion csv mappings
Update policy not firingVerify function runs standalone; check .show table T policy update
Forbidden (403) on management commandsRequest admin or ingestor database role
Materialized view stuckCheck .show materialized-view MV statistics; may need .disable/.enable
OneLake ingest auth errorAdd ;impersonate to abfss:// URI

Agentic Workflows

Exploration Before Authoring

Always check for explicit intent before doing anything:

Step 0 → Is the request specific? Does it name a table, operation, and/or schema?
         → NO  → Ask: "What would you like to set up? Options: create tables,
                  configure policies, set up ingestion mappings, create materialized views."
                  STOP — do not proceed until user specifies.
         → YES → Continue to Step 1.
Step 1 → .show tables details                        // what exists?
Step 2 → .show table <TABLE> schema as json          // current columns
Step 3 → .show table <TABLE> policy retention        // current policies
Step 4 → Plan changes (create-merge, alter, etc.)
Step 5 → Execute changes
Step 6 → Verify: .show table <TABLE> schema as json  // confirm changes
Script Generation Workflow
Step 1 → Understand requirements from user
Step 2 → Generate KQL management commands
Step 3 → Save to .kql file
Step 4 → Deploy via az rest (one command at a time)
Step 5 → Verify deployed state matches intent

Examples

Example 1: Create Table with Policies and Mapping
# Create table
read -r -d '' CSL <<'KQL' || true
.create-merge table SensorData (Timestamp: datetime, DeviceId: string, Temperature: real, Humidity: real, Location: dynamic)
KQL
jq -n --arg db "${DB}" --arg csl "${CSL}" '{db: $db, csl: $csl}' > /tmp/kql_body.json

Execute /tmp/kql_body.json — see Execute KQL Command

# Set retention
read -r -d '' CSL <<'KQL' || true
.alter table SensorData policy retention '{"SoftDeletePeriod":"90.00:00:00","Recoverability":"Enabled"}'
KQL
jq -n --arg db "${DB}" --arg csl "${CSL}" '{db: $db, csl: $csl}' > /tmp/kql_body.json

Execute /tmp/kql_body.json — see Execute KQL Command

# Set caching
read -r -d '' CSL <<'KQL' || true
.alter table SensorData policy caching hot = 7d
KQL
jq -n --arg db "${DB}" --arg csl "${CSL}" '{db: $db, csl: $csl}' > /tmp/kql_body.json

Execute /tmp/kql_body.json — see Execute KQL Command

# Create JSON mapping
read -r -d '' CSL <<'KQL' || true
.create table SensorData ingestion json mapping 'SensorJsonMapping' '[{"column":"Timestamp","path":"$.ts","datatype":"datetime"},{"column":"DeviceId","path":"$.deviceId","datatype":"string"},{"column":"Temperature","path":"$.temp","datatype":"real"},{"column":"Humidity","path":"$.humidity","datatype":"real"},{"column":"Location","path":"$.location","datatype":"dynamic"}]'
KQL
jq -n --arg db "${DB}" --arg csl "${CSL}" '{db: $db, csl: $csl}' > /tmp/kql_body.json

Execute /tmp/kql_body.json — see Execute KQL Command

Example 2: ETL with Update Policy
// 1. Target table
.create-merge table ParsedLogs (Timestamp: datetime, Level: string, Message: string, Source: string)

// 2. Transform function
.create-or-alter function ParseRawLogs() {
    RawLogs
    | extend J = parse_json(RawMessage)
    | project
        Timestamp = todatetime(J.timestamp),
        Level = tostring(J.level),
        Message = tostring(J.message),
        Source = tostring(J.source)
}

// 3. Attach update policy
.alter table ParsedLogs policy update
@'[{"IsEnabled":true,"Source":"RawLogs","Query":"ParseRawLogs()","IsTransactional":true}]'

Agent Integration Notes

  • This skill covers authoring operations — creating/altering database objects and ingesting data.

  • For read-only queries and data exploration, switch to the eventhouse-cli consumption mode.

  • For cross-workload orchestration, delegate to the FabricDataEngineer agent.

  • All management commands require elevated database roles (Admin or Ingestor).

Requirements and compatibility

Works with: Claude Code, Claude apps, OpenAI Codex, GitHub Copilot, Cursor, Gemini CLI, Kiro, OpenCode, Goose, Amp, Roo Code, JetBrains Junie, Factory Droid, Devin (ex-Windsurf), Mistral Vibe, OpenHands, Letta

“Works with” means the tool reads the open Agent Skills format; a skill written for one tool may use features others lack.

Security

What our scanner noticed· Low risk

  • Scripts that your agent may run

This skill includes scripts that your agent may run. Read them before you use it.

A static scan is not a guarantee. Skills are third-party content: review a skill before you let an agent use it.

Attribution and licence

Published by
Microsoft
Publisher
Microsoft (Official)
Source
microsoft/skills-for-fabric · skills/eventhouse-cli · commit f8fac11
Licence
MIT License (redistribution allowed)
  • LICENSE (added by AutoGineer from LICENSE)
  • ATTRIBUTION.md (written by AutoGineer)
Synced
Also distributed in
microsoft/skills-for-fabric

AutoGineer is not affiliated with or endorsed by Microsoft or Microsoft.

Download ZIP

The ZIP holds the skill’s files, its licence files and an ATTRIBUTION.md. Download eventhouse-cli.zip