create-dive
Create MotherDuck Dives
Use this skill when the user needs a persistent, shareable Dive rather than a one-off chart or a full customer-facing analytics application.
Source Of Truth
- Prefer current MotherDuck Dive docs first.
- If MotherDuck MCP is available, call
get_dive_guidebefore generating, saving, or updating a Dive. - Keep theming, local-preview, code-management, and embedding guidance aligned with the current Dive docs.
Default Posture
- Validate the underlying SQL and schema first with
exploreandquery. - Keep Dive queries fully qualified and SQL-heavy; let React handle presentation, not data reshaping.
- When a Dive uses shared databases in
REQUIRED_DATABASES, always suffix thealias(e.g._share) so it cannot collide with an existing database name. Never setaliasto conflict with an existing database name. - Start from a named theme direction such as
Corporate Dashboard,Tufte Minimal, orFT Salmoninstead of vague visual prompts. - Prefer one query per visual section rather than one giant cross-purpose query.
- Preview locally before saving when the environment supports it.
- Treat embedded Dives as a lightweight read-only surface; for per-customer serving and policy control, move to
build-cfa-app.
More from motherduckdb/agent-skills
motherduck-security-governance
Explain MotherDuck security, governance, and access-control patterns. Use when a security_compliance_owner, technical_owner, or application_builder is asking about residency, access boundaries, service accounts, isolation, sharing, or governance posture.
56motherduck-duckdb-sql
DuckDB SQL reference for MotherDuck. Use when you need exact DuckDB syntax, function behavior, supported MotherDuck SQL features, or to resolve whether PostgreSQL-oriented SQL will fail on MotherDuck.
55motherduck-build-cfa-app
Design a MotherDuck-backed customer-facing analytics app. Use when building embedded or product analytics for external users and the decision depends on per-customer isolation, backend routing, service-account boundaries, read scaling, or Hypertenancy-style patterns.
54motherduck-build-data-pipeline
Design an end-to-end MotherDuck pipeline. Use when choosing raw, staging, and analytics boundaries, bulk ingestion paths, transformation sequencing, publication targets, or whether DuckLake is actually required.
54motherduck-ducklake
Decide when DuckLake is the right MotherDuck storage pattern. Use when evaluating fully managed DuckLake, BYOB, own-compute DuckLake access, data inlining, object-storage layout, or file-aware maintenance instead of native MotherDuck storage.
54motherduck-create-dive
Create, edit, manage, share, or embed MotherDuck Dives. Use when the work involves Dive authoring, live React + SQL components, MCP get_dive_guide, useSQLQuery, local preview, version history, Dives-as-code, required resources, team sharing, or embedded Dive sessions.
54