plugin

daft — Claude Code plugin: what it does, how to install

Q: How do I install daft in Claude Code?

Run `/plugin marketplace add anthropics/claude-plugins-community` to register the marketplace. Next, execute `/plugin install daft@claude-community`.

Adds Claude Code capabilities for the Daft data engine. It assists with UDF performance tuning, documentation lookup, and Ray scaling.

Install into Claude Code · Plugin marketplace All methods and platforms ↓

/plugin marketplace add anthropics/claude-plugins-community
/plugin install daft@claude-community

Open on GitHub

What daft does

This plugin equips Claude Code with skills targeting the Daft dataframe framework, built specifically for AI pipelines and multimodal datasets. The model gets direct context on navigating official documentation to find relevant APIs.

It guides user-defined function (UDF) optimization and handles multi-node compute setups via Ray. Distributed execution relies directly on Ray, meaning cluster workloads require an active Ray setup.

What it helps with

tune execution speed for Python UDFs in Daft
scale data pipelines across nodes with Ray
search official Daft documentation for API patterns

Example prompts for the agent

How can I tune my Daft UDF to run faster on image data?
Configure distributed execution for Daft using Ray.
Check Daft documentation for reading multimodal Parquet files.

How to install daft

Plugin marketplace

/plugin marketplace add anthropics/claude-plugins-community

/plugin install daft@claude-community

Step by step, with commands for each platform: Claude Code plugins: marketplaces, installing, and building your own

Questions about daft

How do I install daft in Claude Code?

Run /plugin marketplace add anthropics/claude-plugins-community to register the marketplace. Next, execute /plugin install daft@claude-community.

What does the daft plugin do?

It teaches Claude Code how to navigate Daft docs, optimize UDF throughput, and manage data processing for AI pipelines.

Which distributed runner does Daft use here?

Distributed scaling in this plugin is built around Ray cluster integrations.

Source text

SKILL.md · Apache-2.0

Skills for working with Daft, a high-performance data engine for AI and multimodal workloads. Includes UDF performance tuning, distributed scaling with Ray, and documentation navigation.

Similar plugins

All plugins for Data and databases →

firebase official Google Wires Google Firebase into Claude over MCP, so the agent can work with Firestore, auth, cloud functions, hosting and storage from your project. ★ 44,635repo stars power-platform-architect official Tim Hanewich Turns requirements and meeting transcripts into Power Platform architecture recommendations. Use it when you need component choices and a clear data-flow description. ★ 40,418repo stars power-bi-development official Awesome Copilot Community This plugin adds prompts and agents for DAX, data models, performance troubleshooting, and report design. Use it while building or reviewing Power BI work. ★ 40,418repo stars partners official Awesome Copilot Community Collection of partner agents for GitHub Copilot covering Arm migration, Neon database branching, Terraform workflows, and incident triage. ★ 40,418repo stars openapi-to-application-csharp-dotnet official Awesome Copilot Community Generate a .NET application from an OpenAPI specification, with ASP.NET Core scaffolding and controller generation. Use it to start an application from an API definition. ★ 40,418repo stars dataverse-sdk-for-python official Awesome Copilot Community Four slash commands that generate Python code for Microsoft Dataverse, from a quickstart with CRUD to production patterns with error handling. ★ 40,418repo stars