Motherduck Share Data logo

Motherduck Share Data

Organization
motherduckdb
motherduck-share-data

Create, consume, or manage MotherDuck data shares, including audience grants, table filters, and refresh policy.

Overview

Publishermotherduckdb
Repositoryagent-skills
Skill namemotherduck-share-data
Stars
62
Forks
6
Bundled files
1
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 1 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by motherduckdb on GitHub. Read the source before you install it.

Installation

Install the Motherduck Share Data AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/motherduckdb/agent-skills/tree/main/plugins/motherduck-skills-claude/skills/motherduck-share-data
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/motherduckdb/agent-skills.git /tmp/agent-skills
mkdir -p .claude/skills
cp -r /tmp/agent-skills/plugins/motherduck-skills-claude/skills/motherduck-share-data .claude/skills/motherduck-share-data
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Motherduck Share Data in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Motherduck Share Data on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Motherduck Share Data is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Share Data with MotherDuck

Source Of Truth

  • Prefer the current MotherDuck sharing docs and SQL reference first.
  • If the MotherDuck MCP ask_docs_question feature is available, use it before falling back to public docs.
  • Keep the sharing model aligned with the documented behavior:
    • zero-copy and metadata-only
    • a database is the share source, with optional table/view filtering through INCLUDE_PATTERN
    • optional secure views that filter rows, drop columns, or mask values for each consumer
    • read-only recipients
    • owner-controlled update and include-pattern policy

Prerequisites

A working connection, the source database, and the intended audience. Reuse these from context; related skills are available for missing setup or SQL details.

Default Posture

  • Prefer ACCESS RESTRICTED plus GRANT READ ON SHARE ... TO ROLE ... for governed internal distribution. Use ACCESS ORGANIZATION only when the caller deliberately wants its legacy organization-wide behavior.
  • Use INCLUDE_PATTERN when every recipient of one share should see the same table/view subset. Create separate shares when audiences need different table subsets; when they differ only by rows, columns, or masked values, one share with secure views can serve them all.
  • Use UPDATE MANUAL when the recipient needs a stable snapshot or versioned delivery.
  • Use ACCESS RESTRICTED or VISIBILITY HIDDEN when distribution should stay tightly controlled.
  • Confirm whether the recipient is an internal user, another organization, or public before choosing access and visibility.
  • For write-heavy publishers, verify the DuckDB client version is one MotherDuck supports before relying on checkpoint or concurrent-write behavior during share-update workflows.
  • INCLUDE_PATTERN selects whole tables and views; never describe it as row- or column-level security. It works on native MotherDuck and fully managed DuckLake sources.
  • For row filtering, hidden columns, or masked values, add secure views to the share: CREATE SHARE ... WITH SECURE VIEW REPLACES <table> AS (SELECT ...) or ALTER SHARE ... ADD [OR REPLACE] SECURE VIEW. Key the policy on current_email() for people, current_user for service accounts, or is_role_in_session('<role>'). One share with secure views can replace a set of per-audience shares. They are plan-gated, and managing them in SQL needs a recent DuckDB client and a read-write token; check the current secure-views docs for both.
  • Write secure views with explicit column lists, never SELECT * (including SELECT * EXCLUDE or SELECT * REPLACE): a new source column would otherwise leak to every consumer, and schema changes break the view. A table the share exposes without a replacing secure view stays fully readable. Combine INCLUDE_PATTERN (which tables) with secure views (which rows and columns) and keep macro schemas and mapping tables outside the pattern.
  • MotherDuck secure views are share policies. DuckDB's own CREATE SECURE VIEW is only an optimizer barrier and restricts nothing.

Workflow

  1. Identify the exact database to publish and who should consume it.
  2. Decide the audience, visible table/view subset, discoverability, and freshness requirements before writing SQL.
  3. Preview and validate any include pattern against the live source catalog, then create the share with explicit access, visibility, update mode, and optional INCLUDE_PATTERN.
  4. If access is restricted, grant users or roles explicitly. If the share is hidden or link-based, distribute the share URL directly.
  5. Read the share back with LIST SHARES or MD_INFORMATION_SCHEMA.OWNED_SHARES; for filtered shares, verify the stored pattern and consumer-visible catalog. For secure views, run SHOW SECURE VIEWS ON SHARE <name> and attach the share as a consumer to check the result; identity functions evaluate as whoever runs the test.
  6. Have recipients ATTACH the shared database and query it read-only.
  7. Use ALTER SHARE ... SET|RESET INCLUDE_PATTERN for table/view scope changes and ALTER SHARE ... ADD|DROP SECURE VIEW for policy changes (not both in one statement). Avoid CREATE OR REPLACE SHARE on a share with secure views; it drops any view it does not restate. If the share uses UPDATE MANUAL, the owner runs UPDATE SHARE and consumers run REFRESH DATABASE when a new snapshot is ready.

For answer, review, or planning requests, return the sharing design and SQL without provisioning. For create, update, grant, or revoke requests, perform the requested in-scope operation and validate the resulting access; ask before public exposure, destructive revocation, or unrelated grants.

References

Read only the reference sections needed for the current task.

  • For who-gets-what design across teams, tenants, or environments, use motherduck-security-governance (references/ACCESS_CONTROL_SETUPS.md).
  • Read references/SHARE_PLAYBOOK.md for the full SQL playbook, role grants, include-pattern rules, secure views, access/update decisions, consumer workflow, and common failure modes.

Related Skills

Load related skills only for missing capabilities; reuse established context.

  • motherduck-connect for MotherDuck authentication and connection setup
  • motherduck-explore for discovering databases, tables, columns, and existing shares
  • motherduck-query for validating share SQL and downstream queries
  • motherduck-duckdb-sql for DuckDB SQL syntax and lookup support
  • motherduck-security-governance for role design and access-boundary review

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Motherduck Share Data AI skill do?

Create, consume, or manage MotherDuck data shares, including audience grants, table filters, and refresh policy.

Why use Motherduck Share Data on TypingMind?

Because you install it once and use it with any model. Motherduck Share Data is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Motherduck Share Data in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/motherduckdb/agent-skills/tree/main/plugins/motherduck-skills-claude/skills/motherduck-share-data. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Motherduck Share Data?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Motherduck Share Data?

As many as you like. As long as a model supports skills, you can use Motherduck Share Data with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Motherduck Share Data AI skill free?

Yes. It is published on GitHub by motherduckdb under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇