voxel51voxel51Officialยท10 Agent Skills Included

fiftyone-skills

Computer vision dataset import, curation, evaluation, and export

Imports, curates, and exports computer vision datasets in FiftyOne using plain-English requests. Finds duplicate images, runs model inference, computes mAP and confusion matrices, and visualizes embeddings without manual scripting. Builds plugins, notebooks, and zoo model integrations, and fixes common FiftyOne setup errors automatically.
npx skills add voxel51/fiftyone-skills --all -g -y
Available:

Tells the AI agent when to trigger each FiftyOne skill, which MCP server and plugins are required, and the exact step-by-step workflow to follow for dataset import, export, evaluation, and curation tasks.

All Skills in This Repository (10)

Pure Emerald Level Indicators
๐Ÿ“ฆ In Repo
voxel51voxel51

fiftyone-embeddings-visualization

Reduce dataset embeddings to 2D or 3D with UMAP, t-SNE, or PCA for visualization in the FiftyOne app.

Official
Intermediate
๐Ÿ“ฆ In Repo
voxel51voxel51

fiftyone-issue-triage

Categorize FiftyOne GitHub issues into fixed, won't fix, not reproducible, or still valid.

Official
Intermediate
๐Ÿ“ฆ In Repo
voxel51voxel51

fiftyone-dataset-inference

Run Zoo model inference on FiftyOne datasets for detection, classification, and embeddings.

Official
Intermediate
๐Ÿ“ฆ In Repo
voxel51voxel51

fiftyone-dataset-import

Automate dataset imports into FiftyOne with automatic format detection.

Official
Intermediate
๐Ÿ“ฆ In Repo
voxel51voxel51

fiftyone-find-duplicates

Finds and deduplicates near-identical images in FiftyOne datasets using similarity scoring and thresholds.

Official
Advanced
๐Ÿ“ฆ In Repo
voxel51voxel51

fiftyone-dataset-export

Export FiftyOne datasets to COCO, YOLO, VOC, CVAT, and CSV formats.

Official
Intermediate
๐Ÿ“ฆ In Repo
voxel51voxel51

fiftyone-model-evaluation

Evaluate model predictions against ground truth using FiftyOne evaluation protocols.

Official
Intermediate
๐Ÿ“ฆ In Repo
voxel51voxel51

fiftyone-voodo-design

Build FiftyOne plugin panels using the VOODO React component library.

Official
Intermediate
๐Ÿ“ฆ In Repo
voxel51voxel51

fiftyone-code-style

Standardize Python code to FiftyOne's official conventions.

Official
Intermediate
๐Ÿ“ฆ In Repo
voxel51voxel51

fiftyone-develop-plugin

Design and implement custom FiftyOne plugins with Python backends and JavaScript frontends.

Official
Advanced

Frequently Asked Questions

FAQPage Schema
How to install fiftyone-skills?โ–ผ

Run `npx skills add voxel51/fiftyone-skills --all -g -y` in your terminal to install all skills globally. Some skills also require the FiftyOne MCP server, installed with `pip install fiftyone-mcp-server`.

How to find duplicate images in a dataset?โ–ผ

Load your dataset in FiftyOne and ask your assistant to find duplicates. The skill computes similarity embeddings, groups near-duplicates, and lets you review them in the FiftyOne App before deleting.

Can it import COCO, YOLO, or Hugging Face datasets?โ–ผ

Yes. The import skill auto-detects COCO, YOLO, VOC, KITTI, CVAT, and Hugging Face Hub formats, including multimodal grouped datasets like autonomous driving scenes.

Does fiftyone-skills work with Claude Code and Cursor?โ–ผ

Yes. All skills follow the universal SKILL.md standard and work in Claude Code, Cursor, Codex, Gemini CLI, and other compatible assistants.

Can I evaluate model predictions without writing code?โ–ผ

Yes. Ask your assistant to evaluate predictions against ground truth, and it computes mAP, precision, recall, and confusion matrices, then opens the results in the FiftyOne App.

Related Repositories in Data & Analytics

View All in Data & Analyticsโ†’