Annotate anything with Potato: text, media, and agents
Potato is a free, open-source tool for annotating text, audio, images, video, interviews, and AI agent traces. You set up each study in one YAML file, run it on your own computer or server, and watch agreement between annotators while the study runs.
Get started in 4 steps
From installation to annotation in minutes.
1.# Install Potato
pip install potato-annotation2.# Configure your task
# Create your config.yaml3.# Launch the server
potato start config.yaml4.# Start annotating
# Open localhost:80001.# Install Potato
pip install potato-annotation2.# In Claude Code, add the skill's marketplace
/plugin marketplace add davidjurgens/potato-skill3.# Install potato-skill
/plugin install potato-skill@potato4.# Describe the study, and the agent designs and builds it
> Label 200 tweets for sentiment, three annotators each# In Codex or Cursor, use this instead of steps 2 and 3
npx skills add davidjurgens/potato-skill --agent codex cursorWhat's in Potato
Built for researchers and annotation teams.
- Agent Evaluation
- Annotate multi-agent teams on a clickable interaction graph, evaluate computer-use, voice, and video agents, calibrate LLM judges against human labels, edit trajectories into SFT and DPO training data, and rank models in an arena, all from YAML.
- 80 Annotation Types
- Radio, multiselect, likert, slider, text, span, best-worst scaling, pairwise, number, multirate, video, image, audio, bounding box, polygon, event, and taxonomy annotation, plus qualitative coding (QDA Mode) with a living codebook, in-vivo codes, memos, and cases.
- AI-Powered
- LLM integration with OpenAI, Claude, and Gemini for hints, keyword highlighting, label suggestions, MACE competence estimation, option highlighting, and diversity ordering.
- Multimedia Support
- Annotate audio with waveforms, images with bounding boxes and polygons, and video with playback controls.
- Active Learning
- Five query strategies (uncertainty sampling, diversity-based selection, BADGE, BALD, and hybrid ensemble), plus an LLM cold start that picks the first instances before any labels exist.
- Zero Code Setup
- Configure everything in YAML. Building an annotation interface takes no programming.
Find your field
Each field page lists the showcase designs and guides for one kind of study.
All use cases- Natural Language Processing
- Computer Vision
- Audio & Speech
- Robotics and Embodied AI
- LLM Evaluation & RLHF
- AI Agent Evaluation
- Psychology & Cognitive Science
- Survey Research
- Qualitative Research
- Computational Social Science
- Linguistics
- Digital Humanities
- Legal Annotation
- Healthcare & Medical
- Sleep and Physiology
- Microscopy and Pathology
- Life Sciences
- Chemistry
- Ecology
- Astronomy
Guides to annotation practice
Explainers that answer a question about annotation without assuming a tool, then use Potato as the worked example.
Browse all guidesBrowse annotation designs
460+ configurations you can download and run.
Ready to start annotating?
Potato is free and open source. Install it with pip and run your first task locally.