🎣 BAIT — Biotic AI Toolkit
Biotic goes agentic
BAIT teaches your AI coding agent (Claude Code, Codex, Cursor, Mistral Vibe, …) to work with the IMR Biotic database. Most Biotic data are open; some subsets and outputs are restricted or sensitive.
Ask in plain language:
“Set up the Biotic database.” “What is the largest cod in the database?” “Make a map of cusk caught on the EggaN surveys.” “Make a maturity ogive for ling using Coastal survey data.”
…and your agent writes the R, runs it locally, and shows you the result.
What it does
- Installs & maintains the Biotic database (BioticExplorerServer).
- Connects from R with tidyverse syntax (read-only DuckDB).
- Answers data questions by translating them into queries.
- Makes maps, plots, life-history analyses, and dashboards with ggplot2, ggOceanMaps, leaflet, ggFishPlots, and BioticExplorer/Shiny.
It is not a trained model
BAIT is a knowledge pack, not a model trained on Biotic records. Local R/DuckDB execution does not guarantee that prompts, tool inputs, logs, or outputs stay on-device or are excluded from provider processing, retention, or training. See Privacy.
Get going
- New here? → Getting started
- Want ideas? → Example queries
- Worried about data? → Privacy (read this first)