plugin
Open on GitHub
fiftyone — Claude Code plugin: what it does, how to install
A set of skills for computer vision workflows: import datasets, detect duplicates, run inference, and inspect model predictions.
Install into Claude Code · Plugin marketplace
All methods and platforms ↓
/plugin install fiftyone@claude-plugins-official
/plugin marketplace add anthropics/claude-plugins-community
/plugin install fiftyone@claude-community
/plugin marketplace add anthropics/claude-plugins-officialWhat fiftyone does
The plugin gives Claude Code packaged workflows for FiftyOne and computer vision. Skills cover dataset import and export, duplicate detection, Zoo model inference, model evaluation, embedding visualization, and end-to-end dataset curation. Other workflows create FiftyOne plugins, Jupyter notebooks, Data Lens connectors, remote Zoo model integrations, and Python code that follows FiftyOne conventions.
Skills marked for MCP use require the FiftyOne MCP Server to interact with datasets and run operators. Install it with pip install fiftyone-mcp-server, then configure the AI tool with the same Python environment where that package was installed. This environment requirement is a real setup constraint for MCP-backed workflows.
The package also includes workflows for VOODO React interfaces, GitHub issue triage, troubleshooting, plugin evaluation, and driving the FiftyOne App through Playwright MCP. Some skills work without MCP, so the available workflow depends on what you ask Claude Code to do.
What it helps with
- import a COCO dataset into FiftyOne
- remove duplicate images from a dataset
- run detection or classification with a Zoo model
- measure predictions with mAP, precision, and recall
- inspect embedding clusters and outliers
Example prompts for the agent
- Import this COCO dataset into FiftyOne and prepare it for analysis.
- Find duplicate images in my dataset and remove them.
- Run a Zoo model, evaluate its predictions, and show the confusion matrix.
How to install fiftyone
Plugin marketplace
/plugin install fiftyone@claude-plugins-official/plugin marketplace add anthropics/claude-plugins-community/plugin install fiftyone@claude-community/plugin marketplace add anthropics/claude-plugins-officialStep by step, with commands for each platform: Claude Code plugins: marketplaces, installing, and building your own
Questions about fiftyone
How do I install fiftyone in Claude Code?
Run these steps in order: /plugin install fiftyone@claude-plugins-official, /plugin marketplace add anthropics/claude-plugins-community, /plugin install fiftyone@claude-community, /plugin marketplace add anthropics/claude-plugins-official.
What does the fiftyone plugin do?
It adds guided workflows for dataset import and export, duplicate discovery, model inference, evaluation, embedding analysis, and dataset curation. It also covers plugin development, notebooks, troubleshooting, and related FiftyOne tasks.
What is required for the MCP-based fiftyone skills?
Install the FiftyOne MCP Server with pip install fiftyone-mcp-server. Claude Code must use the same Python environment that contains the server.
Source text
SKILL.md · Apache-2.0
# FiftyOne Skills
<div align="center">
<p align="center">
<!-- prettier-ignore -->
<img src="https://user-images.githubusercontent.com/25985824/106288517-2422e000-6216-11eb-871d-26ad2e7b1e59.png" height="55px">
<img src="https://user-images.githubusercontent.com/25985824/106288518-24bb7680-6216-11eb-8f10-60052c519586.png" height="50px">
</p>
**Expert workflows for computer vision powered by AI assistants**
[](LICENSE)
[](https://github.com/voxel51/fiftyone)
[](https://github.com/voxel51/fiftyone-mcp-server)
[](https://github.com/voxel51/fiftyone-skills/actions/workflows/validate.yml)
[](https://discord.gg/fiftyone-community)
[](https://huggingface.co/Voxel51)
[](https://voxel51.com/blog)
[](https://share.hsforms.com/1zpJ60ggaQtOoVeBqIZdaaA2ykyk)
[](https://www.linkedin.com/company/voxel51)
[](https://x.com/voxel51)
[](https://medium.com/voxel51)
[Documentation](https://docs.voxel51.com) · [MCP Server](https://github.com/voxel51/fiftyone-mcp-server) · [FiftyOne Plugins](https://github.com/voxel51/fiftyone-plugins) · [Discord](https://discord.gg/fiftyone-community)
</div>
## What are Skills?
Skills are packaged workflows that teach AI assistants to perform complex computer vision tasks autonomously. Combined with the [FiftyOne MCP Server](https://github.com/voxel51/fiftyone-mcp-server), you can find duplicates, run inference, and explore datasets using natural language.
```
"Find and remove duplicate images from my dataset"
"Import this COCO dataset and run object detection"
"Visualize my embeddings and identify outliers"
```
Skills bridge the gap between natural language and FiftyOne's 80+ operators, providing step-by-step guidance that AI assistants follow to complete complex workflows.
## Available Skills
| Skill | Description | MCP |
|-------|-------------|-----|
| 📥 [**Dataset Import**](skills/fiftyone-dataset-import/SKILL.md) | Universal import for all media types, label formats, multimodal groups, MCAP recordings, LeRobot robot-learning episodes, and Hugging Face Hub | Yes |
| 📤 [**Dataset Export**](skills/fiftyone-dataset-export/SKILL.md) | Export datasets to COCO, YOLO, VOC, CVAT, CSV, Hugging Face Hub, and more | Yes |
| 🔍 [**Find Duplicates**](skills/fiftyone-find-duplicates/SKILL.md) | Find and remove duplicate images using brain similarity | Yes |
| 🤖 [**Dataset Inference**](skills/fiftyone-dataset-inference/SKILL.md) | Run Zoo models for detection, classification, segmentation, embeddings | Yes |
| 📈 [**Model Evaluation**](skills/fiftyone-model-evaluation/SKILL.md) | Compute mAP, precision, recall, confusion matrices, analyze TP/FP/FN | Yes |
| 📊 [**Embeddings Visualization**](skills/fiftyone-embeddings-visualization/SKILL.md) | Visualize datasets in 2D, find clusters, identify outliers | Yes |
| 🔌 [**Develop Plugin**](skills/fiftyone-develop-plugin/SKILL.md) | Create custom FiftyOne plugins (operators and panels) | — |
| 🎨 [**VOODO Design**](skills/fiftyone-voodo-design/SKILL.md) | Build UIs with VOODO React components and design tokens | — |
| 📝 [**Code Style**](skills/fiftyone-code-style/SKILL.md) | Write Python code following FiftyOne's official conventions | — |
| 📓 [**Create Notebook**](skills/fiftyone-create-notebook/SKILL.md) | Create Jupyter notebooks: getting-started guides, tutorials, recipes, ML pipelines | — |
| 🏷️ [**Issue Triage**](skills/fiftyone-issue-triage/SKILL.md) | Triage GitHub issues: validate status, categorize, generate responses | — |
| 🧹 [**Dataset Curation**](skills/fiftyone-dataset-curation/SKILL.md) | End-to-end curation: quality checks, annotation audit, duplicates, class distribution, splits | Yes |
| 🔧 [**Troubleshoot**](skills/fiftyone-troubleshoot/SKILL.md) | Fix common issues: dataset persistence, App connection, MongoDB errors, codecs, performance | — |
| 🛡️ [**Eval Plugin**](skills/fiftyone-eval-plugin/SKILL.md) | Evaluate plugins for quality, security, and agent-readiness. Produces a structured report | — |
| 🧩 [**Zoo Remote Model**](skills/fiftyone-zoo-remote-model/SKILL.md) | Build remote model zoo integrations that work with `register_zoo_model_source` and `dataset.apply_model` | — |
| 🔌 [**Generate Data Lens Connector**](skills/fiftyone-generate-data-lens-connector/SKILL.md) | Generate a Data Lens connector from an external database schema (PostgreSQL, BigQuery, MySQL, etc.) | — |
| 🎭 [**App Playwright**](skills/fiftyone-app-playwright/SKILL.md) | Drive the FiftyOne App via the Playwright MCP for operator/plugin verification, demos, and end-to-end UI automation | — |
| 📚 [**SDK Guidance**](skills/fiftyone-sdk-guidance/SKILL.md) | Answer FiftyOne SDK and docs questions with live documentation search, or provide the SDK path when no operator exists | — |
## Quick Start
### Step 1: Install Skills
**Universal Installer** (Recommended):
```bash
curl -sL skil.sh | sh -s -- voxel51/fiftyone-skills
```
Interactive prompts let you select skills, agents, and install scope (project or global).
Supported agents: Claude Code, Cursor, Codex, OpenCode, GitHub Copilot, Amp, Antigravity, Roo Code, Kilo Code, Goose
**Claude Code:**
```bash
# Register the skills marketplace
/plugin marketplace add voxel51/fiftyone-skills
# Install a skill
/plugin install fiftyone-find-duplicates@fiftyone-skills
```
**Gemini CLI:**
```bash
gemini extensions install https://github.com/voxel51/fiftyone-skills.git --consent
```
### Step 2: Use It
```
"Write a FiftyOne plugin that displays model confidence"
"Write Python code following FiftyOne conventions"
```
Your AI assistant will automatically load the skill instructions and execute the workflow.
### Step 3: Set Up MCP Server (Optional)
Skills marked with **MCP** in the [table above](#available-skills) require the [FiftyOne MCP Server](https://github.com/voxel51/fiftyone-mcp-server) to interact with datasets and run operators.
```bash
pip install fiftyone-mcp-server
```
> **⚠️ Important:** Make sure to use the same Python environment where you installed the MCP server when configuring your AI tool. If you installed it in a virtual environment or conda environment, you must activate that environment or specify the full path to the executable.
Then configure your AI tool:
<details>
<summary><b>Claude Code</b> (Recommended)</summary>
```bash
claude mcp add fiftyone -- fiftyone-mcp
```
</details>
<details>
<summary><b>Claude Desktop</b></summary>
Edit `~/Library/Application Support/Claude/claude_desktop_config.json`:
```json
{
"mcpServers": {
"fiftyone": {
"command": "fiftyone-mcp"
}
}
}
```
</details>
<details>
<summary><b>Cursor</b></summary>
[](cursor://anysphere.cursor-deeplink/mcp/install?name=fiftyone&config=eyJjb21tYW5kIjoiZmlmdHlvbmUtbWNwIn0)
Add to `~/.cursor/mcp.json`:
```json
{
"mcpServers": {
"fiftyone": {
"command": "fiftyone-mcp"
}
}
}
```
</details>
<details>
<summary><b>VSCode</b></summary>
[](https://insiders.vscode.dev/redirect/mcp/install?name=fiftyone&config=%7B%22command%22%3A%22fiftyone-mcp%22%7D)
Add to `.vscode/mcp.json`:
```json
{
"servers": {
"fiftyone": {
"command": "fiftyone-mcp"
}
}
}
```
</details>
Once configured, you can use MCP-dependent skills:
```
"Find and remove duplicate images from my dataset"
"Import this COCO dataset and run object detection"
```
## Skill Structure
Each skill follows the [Agent Skills](https://agentskills.io) specification:
```
skills/
└── fiftyone-find-duplicates/
└── SKILL.md # Instructions for AI
```
**SKILL.md format:**
```markdown
---
name: skill-name
description: When to use this skill
---
# Overview
What this skill does
# Prerequisites
Required setup
# Key Directives
ALWAYS/NEVER rules for AI
# Workflow
Step-by-step instructions
# Troubleshooting
Common errors and solutions
```
## Contributing
We welcome contributions! Whether you want to add a new skill, improve an existing one, or help with integrations and tooling — there's a place for you here.
See **[CONTRIBUTING.md](CONTRIBUTING.md)** for the full guide, including skill structure requirements, the quality bar, and how to test your skill before submitting.
Looking for ideas? Browse issues labeled [`help wanted`](https://github.com/voxel51/fiftyone-skills/labels/help%20wanted) or [`good first issue`](https://github.com/voxel51/fiftyone-skills/labels/good%20first%20issue), or check the [project milestones](https://github.com/voxel51/fiftyone-skills/milestones) for planned work.
## Feedback
Help us improve FiftyOne Skills!
**Just ask your AI assistant:**
```
"Help me submit feedback about [your issue]"
```
The agent will automatically gather session context, environment info, and can submit directly via `gh` CLI or generate content to paste at **[Submit Feedback](https://github.com/voxel51/fiftyone-skills/issues/new?template=skill-feedback.yml)**
## Resources
| Resource | Description |
|----------|-------------|
| [FiftyOne Docs](https://docs.voxel51.com) | Official documentation |
| [FiftyOne MCP Server](https://github.com/voxel51/fiftyone-mcp-server) | MCP server for AI integration |
| [FiftyOne Plugins](https://github.com/voxel51/fiftyone-plugins) | Official plugin collection |
| [Agent Skills Spec](https://agentskills.io) | Skills format specification |
| [PyPI Package](https://pypi.org/project/fiftyone-mcp-server/) | MCP server on PyPI |
| [Discord Community](https://discord.gg/fiftyone-community) | Get help and share ideas |
## 🧡 Community
Join the FiftyOne community to get help, share your skills, and connect with other users:
- **Discord**: [FiftyOne Community](https://discord.gg/fiftyone-community)
- **GitHub Issues**: [Report bugs or request features](https://github.com/voxel51/fiftyone-skills/issues)
---
<div align="center">
Copyright 2017-2026, Voxel51, Inc. · [Apache 2.0 License](LICENSE)
</div>
© voxel51 · Apache-2.0 · voxel51/fiftyone-skills
Similar plugins
All plugins for AI and ML →
discernment-nudge
official This plugin adds two or three short follow-up questions after a substantive answer. Use it to check facts, probe reasoning, and spot missing context.
★ 184,013repo stars
academy-guide
official Recommends Claude Academy courses, tutorials, and use cases when you ask how to use Claude. It suits people learning Claude features through real tasks.
★ 184,013repo stars
claude-api
official This plugin adds Claude API and SDK guidance for building LLM-powered applications. Install it when your project needs a Claude integration.
★ 184,013repo stars
example-skills
official A collection of Claude skill examples for web testing, MCP work, design and documents. Use it to inspect skill structure or start a custom skill from a template.
★ 184,013repo stars
mem0 Adds persistent memory to Claude Code so your agent retains project context across terminal sessions.
★ 67,869repo stars
promptfoo-evals Guides Claude Code in generating and maintaining promptfoo test suites for LLM apps. Covers hallucination checks, JSON schema validation, and CI setup.
★ 26,477repo stars
