Language:JapaneseEnglish

preset-cli-skills Claude Code Plugin

Author
Preset
Category
Development
Topics
Data Engineering & Analytics · Cloud, Deployment & CI/CD · Claude Code Customization & Workflow
First cataloged
2026-07-09 (UTC)
Explanation last updated
2026-09-23 (UTC)
Source (GitHub) last updated
2026-10-05 (UTC) (10 days ago)

The explanation below is AI-generated. Please verify it against the sources.

preset-cli-skills is an agent plugin that supplies skill definitions for driving the sup command-line tool (distributed as the superset-sup PyPI package) to manage Preset and Apache Superset workspaces through shell commands, scripts, and CI/CD pipelines. According to its README, it contains two skills: one covering non-destructive reads and exports of dashboards, charts, datasets, databases, and queries plus ad-hoc SQL, and another covering state-changing operations such as pushing charts, dashboards, or datasets and syncing assets across workspaces. The mutating skill requires a preview or diff of the change and explicit user confirmation before execution. The README specifies that this package should be used only for explicit sup/CLI workflows, and directs users toward separate sibling packages, preset-mcp-skills for MCP-based interactions and preset-api-skills for direct HTTP or SDK calls. It also lays out credential-handling guidance, such as authenticating with environment variables instead of exposing API tokens on the command line.

About the service

Preset is a managed cloud business intelligence service built on the open-source Apache Superset project, offering dashboard creation, embedded analytics, and workspace management. This plugin lets an agent operate on Preset workspaces indirectly through the sup command-line tool rather than through a direct API or MCP connection.

What you can do with preset-cli-skills

  • Install and authenticate the sup CLI using environment-variable-based API tokens
  • List and select Preset/Superset workspaces and output results in JSON, CSV, or YAML
  • Run ad-hoc SQL and read or export dashboards, charts, datasets, databases, and saved queries
  • Push charts, dashboards, or datasets, or synchronize assets across workspaces, with mandatory preview and confirmation steps
  • Follow guardrails for redacting tokens and previewing destructive commands before they run

Sources

Back to list