agentmorrisagentmorrisCommunityยท2 Agent Skills Included

agentmorrispublic

Map wildlife camera-trap labels across datasets and taxonomies

Builds CSV mappings that align animal category labels between camera-trap datasets and species classification models. Handles messy real-world cases like many-to-one species collapses, regional name differences, and taxonomic rank mismatches. Validates every mapping against the official SpeciesNet taxonomy so model evaluation results stay accurate.
npx skills add agentmorris/agentmorrispublic --all -g -y

All Skills in This Repository (2)

Pure Emerald Level Indicators

Frequently Asked Questions

FAQPage Schema
How to install agentmorrispublic skills?โ–ผ

Run `npx skills add agentmorris/agentmorrispublic --all -g -y` in your terminal to install all skills in this repository globally.

How to map species labels between two camera-trap datasets?โ–ผ

The animal-category-mapping skill reads the category lists from two COCO-format JSON files and produces a best-effort source-to-target CSV, flagging judgment calls in a notes column for your review.

How to restrict SpeciesNet predictions to my own species list?โ–ผ

The speciesnet-taxonomy-mapping skill turns your study-area label list into a mapping CSV that MegaDetector's restrict_to_taxa_list() function uses to collapse SpeciesNet's ~3000 categories down to your labels.

Do these skills handle ambiguous species names?โ–ผ

Yes. They use region context, taxonomy lookups, and web search to resolve naming mismatches, and they leave genuinely ambiguous mappings blank with explanatory notes rather than guessing.

Do I need to be a taxonomist to use these skills?โ–ผ

No. You provide your label list and region, and the skill applies taxonomic knowledge automatically, asking you only about genuine judgment calls before writing the final CSV.

Related Repositories in Data & Analytics

View All in Data & Analyticsโ†’