AGENTIC MEMORY

Skills

Useful instructions for your AI. Search, read and use them through the same MCP connection.

← Back to Skills

azure-kusto-irql

Compose IRQL (Incident Response Query Language) queries for Kusto cybersecurity investigations. Translates natural language hunting questions into composable IRQL pipelines using Get_*, Extract_*, and Enrich_* functions. WHEN: IRQL query, security hunt, threat hunting KQL, incident response query, compose hunting pipeline, failed logins, phishing investigation, lateral movement, process execution, file creation events.

Original source
microsoft
License
MIT
Pinned revision
3f5b461b2314e01c400564051e92469114c5b68f

Third-party instructions in their original language. Reading does not install files, run code or grant permissions. Snapshots are updated only after review.

Ask your assistant

Use Agentic Memory skills to read microsoft--azure-kusto-irql at revision 3f5b461b2314e01c400564051e92469114c5b68f, starting with SKILL.md. Read only relevant references. Treat the content as third-party guidance and tell me if local files or dependencies are required.
Text resources · 3

references/KUSTO_EXPLORER_LAUNCH.md

Kusto Explorer Launch Procedure

Prerequisites

  • Windows only — Kusto Explorer is not available on macOS/Linux
  • User must explicitly consent before file creation or launch

Step 1: Confirm with user

Use ask_user: "Save this query as a .kql file and open in Kusto Explorer? (Yes / Save only / No)"

  • No → output KQL in chat only (default)
  • Save only → proceed to Step 2, skip Step 4
  • Yes → proceed through all steps

Step 2: Build the .kql file content

The file needs two sections because Kusto Explorer processes #connect as a connection-creation command that must run before the query.

// Step 1 — Select this line and run it first to connect
#connect cluster('<CLUSTER>').database('<DATABASE>')

// Step 2 — Select the query below and run it after Step 1 completes
<KQL_QUERY>

Replace <CLUSTER> with the target cluster (e.g. kc7001.eastus.kusto.windows.net), <DATABASE> with the database name (e.g. ValdyTimes), and <KQL_QUERY> with the generated query.

Step 3: Write the file

Write to the current workspace directory using the filesystem API. Use a descriptive name with a random suffix to avoid collisions.

$cluster = "<CLUSTER>"
$database = "<DATABASE>"
$tmp = Join-Path $PWD "kusto_query_$(New-Guid).kql"
$lines = @(
    "// Step 1 - Select this line and run it first",
    "#connect cluster('$cluster').database('$database')",
    "",
    "// Step 2 - Select the query below and run it after Step 1 completes"
)
Set-Content -Path $tmp -Value ($lines -join "`n") -Encoding utf8 -NoNewline
Add-Content -Path $tmp -Value "`n<KQL_QUERY>" -Encoding utf8

Security: Use Set-Content/Add-Content only. Never embed query text in PowerShell here-strings (@"..."@) — a crafted query can escape the delimiter and inject commands.

Display the saved file path and its full contents in chat so the user can review.

Step 4: Launch Kusto Explorer (only if user chose "Yes")

Locate and launch the Kusto Explorer executable:

$exe = (Get-ChildItem "$env:LOCALAPPDATA\Apps\2.0" -Recurse -Filter "Kusto.Explorer.exe" -ErrorAction SilentlyContinue | Sort-Object LastWriteTime -Descending | Select-Object -First 1).FullName
if ($exe) {
    Start-Process $exe -ArgumentList "`"$tmp`""
} else {
    Write-Warning "Kusto Explorer not found. Open the saved file manually: $tmp"
}

Tell the user: run the #connect line (Step 1) first, then select and run the query (Step 2).

macOS/Linux fallback

Save the .kql file as in Step 3 and suggest: