$ loading_
Author and manage Fabric Dataflows Gen2 with preview and refresh via CLI.
The material indicates an open-source prompt/documentation-style skill from a Microsoft GitHub repository with some community adoption, so overall risk is relatively low. It does cover using az/curl to call Fabric write APIs for managing dataflows and connections, so normal caution is warranted for execution, data access, and potential network egress, but no clear red flags such as opaque exfiltration, closed-source behavior, or obvious overreach are present.
The material and objective checks indicate the skill itself does not require additional secrets or environment variables. While the README references az login/token acquisition for Fabric access, that is a normal host CLI authentication flow, and there is no indication of collecting, hardcoding, or exfiltrating credentials.
The skill is explicitly intended to use az/curl against Fabric Items and Connections APIs for create/update/delete/refresh/preview operations, which implies outbound requests to function-related remote services. The material does not list unrelated or unknown external endpoints, and the system check says there is no declared remote host, so this is ordinary egress consistent with the stated functionality.
The README explicitly lists a tool stack of az, jq, base64, and curl and includes runnable bash/PowerShell recipes, indicating typical use will execute local CLI tools and scripts. Such local process execution is inherent to CLI/skill behavior, and the material does not show extra privilege escalation, persistence, or system actions unrelated to the stated purpose.
This skill focuses on write-side management of Fabric Dataflows Gen2, involving workspaces, dataflows, connections, output destinations, and mashup/queryMetadata definitions, so it will handle related configuration and content data as part of normal operation. The material does not show access to local files or system resources beyond that business scope, but its ability to manage connections and destinations means least privilege and scope control should be observed.
The source is an open-source Microsoft-related GitHub repository, making the code/docs auditable, and about 420 stars provide positive community-trust evidence that lowers risk. Although the license is not stated and maintenance status is unknown, and the README contains directive text like running a check-updates step, the available facts do not show closed-source opacity, abandonment, or clear supply-chain red flags.
Copy the install command and let the AI configure it · recommended for beginners
Please install the "dataflows-authoring-cli" skill from askskill: 1. Download https://raw.githubusercontent.com/microsoft/skills-for-fabric/main/skills/dataflows-authoring-cli/SKILL.md 2. Save it as ~/.claude/skills/dataflows-authoring-cli/SKILL.md 3. Reload skills and tell me it's ready
Use dataflows-authoring-cli to create a Fabric Dataflows Gen2 dataflow: first create a data source connection to Azure SQL, then bind the connection to the dataflow, set the output destination to Lakehouse, and return the required commands with parameter explanations.
A step-by-step set of CLI commands to create the connection, bind it, create the dataflow, and configure the Lakehouse output.
Use dataflows-authoring-cli to iterate on a Power Query M script: preview results with executeQuery + customMashupDocument first, then adjust the mashup based on errors until the preview succeeds, and provide the final definition file structure ready to save.
A preview-driven debugging flow, the corrected M script, and the dataflow definition structure ready for saving.
Use dataflows-authoring-cli to update an existing Fabric Dataflow query definition, change its output destination to Warehouse, then trigger a parameterized refresh and list the corresponding commands, example parameters, and key caveats.
Command examples and execution notes for updating the dataflow, changing the output destination, and triggering a parameterized refresh.
Update Check — ONCE PER SESSION (mandatory) The first time this skill is used in a session, run the check-updates skill before proceeding.
- GitHub Copilot CLI / VS Code: invoke the
check-updatesskill.- Claude Code / Cowork / Cursor / Windsurf / Codex: compare local vs remote package.json version.
- Skip if the check was already performed earlier in this session.
CRITICAL NOTES
- To find the workspace details (including its ID) from workspace name: list all workspaces and, then, use JMESPath filtering
- 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
This skill (SKILL.md)
| Section | Notes |
|---|---|
| Tool Stack | az + jq + base64 + curl |
| Connection | Workspace/dataflow ID discovery |
| Agentic Workflows | Start here. A: create end-to-end; B: modify existing; C: preview loop |
| MUST DO / AVOID / PREFER | Authoring rules |
| Troubleshooting | Symptom → fix table |
| Examples | Runnable bash + PowerShell recipes |
| Output Expectations | Response conventions |
References (in references/)
| File | When to read |
|---|---|
| authoring-cli-quickref.md | One-liner recipes, status enums, base64 helpers, connection-binding quick patterns |
| authoring-script-templates.md | Full bash + PowerShell templates; end-to-end smoke test; LRO polling pattern |
| connection-management.md | List/create/inspect connections; supportedConnectionTypes; resolve ClusterId; ID format cheat sheet |
| mashup-preview.md | executeQuery contract: bootstrap branch, auto-wrap rule, hard avoid for unbounded preview |
| output-destinations.md | Output destination patterns: Lakehouse Table, Lakehouse Files, Warehouse, ADX, Azure SQL. DataDestinations annotation, hidden query, loadEnabled rules, connection limitations |
Common refs (in ../../common/)
| File | When to read |
|---|---|
| COMMON-CLI.md | az login, token acquisition, az rest, pagination, LRO polling, CLI gotchas. § Finding Workspaces and Items in Fabric is mandatory. |
| COMMON-CORE.md | Fabric topology, environment URLs, authentication, core REST API surface |
| ITEM-DEFINITIONS-CORE.md | Definition envelope; per-item-type payload contracts |
| DATAFLOWS-AUTHORING-CORE.md | Authoring capability matrix; 3-part definition structure; M structure; connection model; ALM / Git integration |
Sister skills
| Skill | Use for |
|---|---|
| dataflows-consumption-cli | Execute persisted queries; ad-hoc read-only customMashupDocument with no intent to persist; Arrow → CSV/pandas conversion; refresh status/history. |
| Tool | Role | Install |
|---|---|---|
az CLI | Primary: Auth (az login), REST API calls (az rest), token acquisition. | Pre-installed in most dev environments |
jq | Parse and manipulate JSON responses and definition payloads. | Pre-installed or trivial |
base64 | Encode/decode definition parts for the REST API. | Built into bash / [Convert]::ToBase64String() in PowerShell |
curl | Alternative to az rest when raw HTTP control is needed. | Pre-installed |
…
Inspect Fabric alerts, notifications, and automated actions via read-only API calls.
Run authoring T-SQL for Fabric warehouses and SQL endpoints from CLI.
Monitor, inspect, and query Fabric Dataflows Gen2 through a read-only CLI.
Query Fabric Eventhouse data, analyze time series, and monitor ingestion health.
Migrate Databricks notebooks, jobs, and paths into Microsoft Fabric.
List, inspect, and validate Microsoft Fabric Eventstream topology and configuration.
Run read-only T-SQL queries on Fabric warehouse and lakehouse data.
Scan, assess, and upgrade Power BI Gen1 dataflows into Gen2.1 copies.
Create, update, and publish Microsoft Fabric Eventstream topologies via REST APIs.
Run DAX queries and inspect Fabric Power BI semantic model metadata.
Analyze Fabric warehouse performance via CLI and get optimization recommendations.
Build Fabric Spark workflows, author notebook code, and manage lakehouse resources.