MotherDuck Skills
22 skills for coding agents working with MotherDuck: connect applications, query real data with DuckDB SQL, build Dives and Flights, and deliver analytics pipelines and apps.
Each skill includes focused instructions and, where useful, detailed references and runnable examples. Installing the skills does not configure a MotherDuck connection or MCP server.
Install
Choose one route for your agent. Run commands beginning with / inside the agent; run the others in your terminal.
| Agent | Install |
|---|---|
| Claude Code / GitHub Copilot CLI | /plugin marketplace add motherduckdb/agent-skills, then /plugin install motherduck-skills@motherduck-skills |
| Codex | codex plugin marketplace add motherduckdb/agent-skills, then open /plugins and install MotherDuck Skills |
| Gemini CLI | gemini extensions install https://github.com/motherduckdb/agent-skills |
| Cursor / VS Code / other supported agents | Use the Skills CLI below |
The Skills CLI needs Git and Node.js 22.20+ (package requirements). No global CLI installation is needed:
npx -y skills add motherduckdb/agent-skills
Select the skills and agents interactively. To install the whole catalog for Cursor:
npx -y skills add motherduckdb/agent-skills --agent cursor --skill '*' --yes --global
Use --agent github-copilot for VS Code/GitHub Copilot, or another supported agent name. Omit --global for a project-scoped install.
See the install matrix for alternatives, updates, verification, and manual installation. Harness support documents the discovery paths and plugin manifests.
Use the Skills
Ask for the task directly, or name a skill from the catalog:
Use motherduck-connect to choose the connection path for this app.
Explore my MotherDuck workspace and build a sales dashboard as a Dive.
Plan a migration to MotherDuck with validation and rollback steps.
For live work, configure MotherDuck MCP or an authenticated database client. Supply credentials through your normal secret or environment setup, never in prompts or committed files.
Start with the skill that matches the task. Reuse established connection and schema context; load supporting references only when needed. The skills favor DuckDB SQL, live schema discovery, native MotherDuck storage by default, and backend credentials for customer-facing apps.
Skills Overview
Utility skills handle individual operations; workflow skills cover specific MotherDuck capabilities; use-case skills combine them for end-to-end work.
| Skill | Layer | Use it for |
|---|---|---|
motherduck-connect |
Utility | Connections, authentication, client selection, and read scaling |
motherduck-cli |
Utility | Terminal queries and file-based Dive/Flight workflows |
motherduck-explore |
Utility | Databases, schemas, shares, and sample data |
motherduck-query |
Utility | Writing, executing, and optimizing analytical SQL |
motherduck-duckdb-sql |
Utility | DuckDB syntax and MotherDuck feature support |
motherduck-rest-api |
Utility | Service accounts, tokens, Ducklings, and embed sessions |
motherduck-load-data |
Workflow | Bulk ingestion from files, storage, or external systems |
motherduck-model-data |
Workflow | Analytical schemas and transformation models |
motherduck-manage-guides |
Workflow | Business definitions and reusable warehouse context |
motherduck-share-data |
Workflow | Share audiences, grants, table filters, and refresh policy |
motherduck-create-dive |
Workflow | Creating, editing, publishing, sharing, or embedding Dives |
motherduck-create-flight |
Workflow | Running and scheduling Python jobs on MotherDuck |
motherduck-design-dive |
Workflow | Dive layout, responsive behavior, themes, and accessibility |
motherduck-ducklake |
Workflow | DuckLake storage choices and operations |
motherduck-security-governance |
Workflow | Permissions, isolation, residency, and compliance |
motherduck-pricing-roi |
Workflow | Costs, plan fit, and ROI |
motherduck-build-cfa-app |
Use-case | Customer-facing analytics with tenant isolation |
motherduck-build-dashboard |
Use-case | A dashboard's analytical story, metrics, and section queries |
motherduck-build-data-pipeline |
Use-case | Ingestion-to-serving pipelines |
motherduck-migrate-to-motherduck |
Use-case | Migration, reconciliation, cutover, and rollback |
motherduck-enable-self-serve-analytics |
Use-case | Governed analytics for internal teams |
motherduck-partner-delivery |
Use-case | Repeatable delivery across client engagements |
Browse the source skills or the machine-readable catalog.
Troubleshooting
- Skill not discovered: check that it is installed for the agent you are using, reload the agent, and name the skill explicitly.
- Live queries fail: verify the MCP server or database connection and its credentials. Skills provide guidance, not account access.
- PostgreSQL-only syntax: use
motherduck-duckdb-sql. PostgreSQL wire compatibility does not change the SQL dialect.
Contributing
Edit source skills under skills/; the plugins/ directories contain generated packages. See CONTRIBUTING.md for checks and package synchronization, skill authoring for conventions, and ARCHITECTURE.md for catalog rules.
Releases use the release process and scripts/bump_version.py; do not edit version numbers by hand.
Resources
License
MIT.