Lightdash skill
Lightdash skills: metrics, charts and dashboards as YAML, branch previews, dbt SQL rules and a safety check before a server upgrade
- 6,230 ★ on the repository
- Last commit
Extended access
Why this level
- The skill steers the agent to `lightdash deploy` and `upload`, which change the project and content on the Lightdash server.
- The `lightdash sql` command runs SQL in your warehouse with the connected user's permissions.
Install
- Works in
Manual install
lightdash install-skillsThe README command. It needs the Lightdash CLI installed. It downloads the skills into the chosen agent's directory: Claude, Cursor or Codex.
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- The skill steers the agent to `lightdash deploy` and `upload`, which change the project and content on the Lightdash server.
- The `lightdash sql` command runs SQL in your warehouse with the connected user's permissions.
This is third-party code. Review the repository files before installing.
What it does
The `skills/` folder of the Lightdash repository holds four skills. The main one, `developing-in-lightdash`, teaches an agent to run an analytics project as code: describe metrics, dimensions and joins in YAML next to dbt models, edit charts and dashboards, export and import content with `lightdash download` and `lightdash upload`, check files with `lightdash lint`, and browse warehouse tables and values through `lightdash warehouse-catalog` and `lightdash sql`. The other three are narrow. `effective-dbt-sql` sets rules for SQL in dbt models: reuse existing fields, prefer CTEs over correlated subqueries, write joins and column lists explicitly. `upgrade-preflight` interprets the answers of `lightdash upgrade-check` and `migrate preflight` before a self-hosted server upgrade. `lightdash-release-review` builds a digest of shipped features over a period, fourteen days by default.
Who it is for. Analytics engineers and dbt data teams who run Lightdash through Git and CI and hand metric, chart and dashboard edits to an agent.
Good fit when
- An agent adds a metric or dimension to a dbt model's YAML and describes a table join.
- A dashboard is exported with `lightdash download`, edited as files and sent back with `lightdash upload`.
- Before merging a branch you need a preview of the changes and a project check with `lightdash preview` and `lightdash validate`.
- An administrator upgrades self-hosted Lightdash and checks the plan against the `lightdash upgrade-check` answer.
- SQL for a dbt model with a metric is written from scratch, and reusing existing fields matters.
Not a fit when
- You work in Lightdash Cloud: the `upgrade-preflight` skill does not fit the cloud, where the vendor manages upgrades.
- The task concerns dbt without Lightdash metadata or raw SQL unrelated to models.
- You are developing the Lightdash application itself: the repository has an `AGENTS.md` for that.
Example request
Add a revenue metric for paid orders to the orders model, update the chart on the sales dashboard and run `lightdash lint` before the preview.Limitations
The skills work through the Lightdash CLI, which needs Node.js 24 or newer and a login to your instance with `lightdash login`. Work with Data Apps and custom chart types is marked enterprise in the skill. The `install-skills` command downloads files from GitHub at the CLI's version tag, so it needs network access and the skill set depends on the installed version. GitHub reports no single licence for the repository, while the CLI package declares MIT.
How to disable. Delete the skill folders from `.claude/skills`, `.cursor/skills` or `.codex/skills` in the project or home directory. The CLI is removed separately with `npm uninstall -g @lightdash/cli`.
FAQ
How do I set up the Lightdash skills for Claude Code?
Install the CLI with `npm i -g @lightdash/cli`, then run `lightdash install-skills`. Files go to `.claude/skills` at the Git repository root; for Cursor and Codex the command uses `.cursor/skills` and `.codex/skills`.
Can I install the skills for all projects at once?
Yes, `install-skills` has a global mode. Files are then written to the home directory: `~/.claude/skills`, `~/.cursor/skills` or `~/.codex/skills`.
Which skills are in the set?
Four: `developing-in-lightdash`, `effective-dbt-sql`, `upgrade-preflight` and `lightdash-release-review`. The first drives the whole project, the others come in by task.
Do the skills work with a self-hosted Lightdash server?
Yes. `lightdash login` saves the login to the right instance by its address, and `upgrade-preflight` is written for self-hosted installs.
How do I update the skills?
Update the CLI and run `lightdash install-skills` again: files are fetched at the CLI's version tag, so a newer version brings a fresh set.
Discussion
No questions or reviews yet.
Related
Skillsdbt Charts
dbt-charts
Declarative YAML on top of SQL for dashboards: you describe how to show data, agents write YAML, people audit. The dbt-charts package and the dct CLI, beta
A dashboards-as-code CLI: an agent describes panels in YAML and TSX, DAC validates them, builds SQL from metrics and serves them
Skills for Claude Code on dbt and Snowflake: models, tests, debugging, migrations, query tuning and handoff to altimate-code
Skill and CLI for tables: cleaning with recorded decisions, level-based statistics, a PDF report and a separate audit.
Streamlit agent skill
developing-with-streamlit
The official Streamlit skill bundled with the library: building, styling, debugging and speeding up apps, installed with streamlit skills
A data visualization skill for agents: turns data into polished interactive HTML charts and full reports with one visual language
Comparison with similar tools
Cards that share the main search query, and how they differ.
| Tool | Official | Authentication | Repository stars | Main difference |
|---|---|---|---|---|
| Lightdash (this card) | yes | 6,230 | ||
| dbt Agent Skills | yes | 762 | official dbt Labs skills for dbt projects; these are Lightdash platform skills: metrics, charts and dashboards as code | |
| dbt MCP | yes | API key | 621 | the dbt Labs MCP server with the Semantic Layer and dbt command execution; these are Lightdash skills and CLI with no separate server |
README in short
The README presents Lightdash as an open Agentic BI platform where metrics, dashboards and AI agents rest on one context layer. For developers it describes a path through Git, the CLI and an MCP server, and installation is possible in the cloud, on your own server through Docker or Helm, and locally for development.
README badge
Original SKILL.md
--- name: developing-in-lightdash description: Use when working with Lightdash YAML files, dbt models with Lightdash metadata, the lightdash CLI (deploy, upload, download, preview, lint, warehouse-catalog, sql, set-warehouse, apps create/preview/validate), or managing charts, dashboards, spaces and access, AI agents, scheduled content, data apps, custom chart types (Chart Studio, `apps create --chart-type`, `--chart-types`), organization Data App themes, data-app external connections, users, groups, custom roles, metrics, and dimensions as code --- # Developing in Lightdash For CLI users and writeback sandboxes working with YAML and dbt files. Build and deploy Lightdash analytics projects. This skill covers the **semantic layer** (metrics, dimensions, joins) and **content** (charts, dashboards). ## When to Use - Working with Lightdash YAML files (charts, dashboards, models as code) - Using the `lightdash` CLI (`deploy`, `upload`, `download`, `preview`, `lint`, `warehouse-catalog`, `sql`) - Defining metrics, dimensions, joins, or tables in dbt or pure Lightdash projects - Creating or editing charts and dashboards as code - Downloading, uploading, or locally developing data apps (enterprise) - Creating, downloading, editing, or uploading custom chart types built in Chart Studio (enterprise) - Creating, editing, migrating, downloading, or uploading organization Data App themes **Don't use for:** Developing the Lightdash application itself (use the codebase AGENTS.md), general dbt work without Lightdash metadata, or raw SQL unrelated to Lightdash models. ## What You Can Do | Task | Commands | References | |------|----------|------------| | Create a pure Lightdash project from warehouse metadata | Use Lightdash or an already-authenticated warehouse CLI to inspect catalog metadata and aggregate profiles | [Creating from a Warehouse Catalog](./resources/creating-from-warehouse-catalog.md) | | Discover warehouse tables and fields | `lightdash warehouse-catalog --json` | [CLI Reference](./resources/cli-reference.md) | | Explore data warehouse values | `lightdash sql` to execute raw sql, read .csv results | [CLI Reference](./resources/cli-reference.md) | | Define metrics & dimensions | Edit dbt YAML or Lightdash YAML | [Metrics](./resources/metrics-reference.md), [Dimensions](./resources/dimensions-reference.md) | | Create or edit charts | UI, Lightdash AI agent, MCP, or YAML via `lightdash download`/`upload` | [Content Workflow](#content-workflow), [Chart Types](#chart-types) | | Add period comparisons | Add PoP additional metrics to chart YAML | [Period over Period](./resources/period-over-period-reference.md) | | Create or edit dashboards | UI, Lightdash AI agent, MCP, or YAML via `lightdash download`/`upload` | [Content Workflow](#content-workflow), [Dashboard Reference](./resources/dashboard-reference.md) | | Manage content as code across project and organization resources | `lightdash download`, `lightdash upload` | [Content as Code](./resources/content-as-code-reference.md) | | Manage data apps as code (enterprise) | `lightdash download --apps <ref>` (one app) or `--include-apps` (all), edit bundle, `lightdash upload --apps <ref>`; local dev via `lightdash apps create/preview/validate` | [Data Apps](#working-with-data-apps-enterprise), [Content as Code](./resources/content-as-code-reference.md) | | Build or edit a custom chart type (enterprise) | `lightdash apps create "<name>" --chart-type` or `lightdash download --chart-types <ref>`, edit `chart-types/<slug>/src/`, `lightdash upload --chart-types <ref>` | [Custom Chart Types](#working-with-custom-chart-types-enterprise) | | Manage organization Data App themes as code | `lightdash download --organization`, edit `themes/<slug>/`, `lightdash upload --organization` | [Data App Themes](./resources/data-app-themes-reference.md) | | Manage data-app external connections (enterprise) | `lightdash download --include-external-connections`, edit YAML, `lightdash upload` | [Content as Code](./resources/content-as-code-reference.md) | | Lint yaml files | `lightdash lint` | [CLI Reference](./resources/cli-reference.md) |