catalog

Plugin catalog from marketplace dataset-catalog in chanzuckerberg/dataset-catalog. Commands read from the repo's manifest — 11 days ago.

Skills (3)

• catalog-read

  Read and analyze the Scientific Dataset Catalog (read-only) — get a dataset by UUID or canonical ID, list a project, browse collections, discover facet values, trace lineage, and aggregate, summarize, compare, or visualize across results. This is the everyday default for known-item lookups and bound

• catalog-register

  Maps a user's dataset metadata onto the latest catalog schema and creates a runnable registration script. Use when asked to map/convert/ingest dataset metadata, fit data to the dataset schema, populate experiment/sample/data_summary/channel fields, or create a script to register/submit a dataset to

• catalog-search

  Advanced high-recall search of the Scientific Dataset Catalog. Expands a term through its synonyms and the OLS ontology hierarchy, runs one search per term, and unions the results by dataset id to round up matches a single query misses (a dataset tagged "hepatic" will not surface for q=liver). Highe

Subagents (1)

• catalog-reader

  Read-only delegate for Scientific Dataset Catalog queries. Use it for dataset searches, lookups by ID or canonical ID, project listings, collection browsing, lineage tracing, and other multi-step queries. The agent runs queries in its own context and returns only a concise result, avoiding raw JSON,

MCP servers (1)

• ols

  connects to: https://www.ebi.ac.uk/ols4/api/mcp

Hooks (1)

• PreToolUse

  "${CLAUDE_PLUGIN_ROOT}/hooks/allow-catalog-reads.py"