Sigma Skills

Sigma AI Coding-Agent Skills

Skills for building and maintaining analytics in Sigma Computing from any AI coding agent.

The skills split along the surfaces of the Sigma REST API: data models, responsive workbooks, fixed-layout reports, and utility skills that orchestrate them. Each skill is a self-contained directory.

**Supported agents** (`generated/` outputs per skill, install via `scripts/install-into-project.sh`):

Agent	Format	How it loads
Claude Code	SKILL.md (canonical)	Symlink the skill into ~/.claude/skills/
Cortex Code (Snowflake)	SKILL.md (same as Claude Code)	Symlink into ~/.snowflake/cortex/skills/ or rely on its ~/.claude/skills/ fallback
Codex (OpenAI CLI)	AGENTS.md	Drop generated/codex/AGENTS.md into project root or ~/.codex/AGENTS.md
Cursor	.cursor/rules/.mdc	Drop generated/cursor/rules/.mdc into /.cursor/rules/
Cline / Roo	.clinerules/.md	Drop generated/cline/.md into /.clinerules/
Continue.dev	.continue/rules/.md	Drop generated/continue/.md into /.continue/rules/

See [Installation](#installation) below for the install helper.

Skills

Skill	Purpose
`sigma-api`	Configure Sigma API credentials and mint short-lived bearer tokens with client credentials or interactive browser OAuth. Prerequisite for skills that call the REST API.
`sigma-data-models`	Author Sigma data models from existing warehouse tables — sources, columns, metrics, relationships, filters, calc columns, CLS. Covers spec shape, discovery, CRUD, validation, and authoring judgment calls. Out of scope: converting from another BI tool's format (use the converter MCP / browser tool).
`sigma-workbooks`	Build, edit, and iterate on Sigma workbook specs — pages, layout, controls, charts (line/bar/area/combo/donut), KPI