publicdata.au

ClaudeCommunityby National Digital

Australian government open data republished as CSV, Excel, JSON, Parquet, SQLite, Arrow, GeoJSON and GeoPackage at URLs that never change. Every release is a dated, immutable version with schema, provenance and a diff. Run by National Digital. No government agency has endorsed it. Tools: - Search datasets: Find datasets on publicdata.au by title, publisher or field name. Returns slug, title, publisher, licence, dataset page and the latest download URL. - Read a dataset: Read one dataset: its Frictionless data package (fields, types, licence, attribution, download URLs) and every version. - List partition files: List the per-value download files of a dataset for one partition field, with the row count and URL of each. query_rows and count_rows do not need them. Use these to download a slice of the data. - Query rows: Return rows from a dataset, filtered across the whole table by the query API. Returns up to limit rows, the number of rows that match, the offset of the next page, the version answered and the attribution string. - Count and sum rows: Count rows in a dataset, or sum, average, min or max a field, grouped by up to three fields and filtered across the whole table by the query API. Returns groups sorted from largest, the number of rows that match, the version answered and the attribution string. - Compare versions: Compare two versions of a dataset by its key: rows added, removed and changed. - Vote for a dataset: Add one vote for a dataset so it is built sooner: a backlog entry, or any catalogue record with an open licence and a download. No identity is recorded. One vote per browser per day. The fields of one dataset by slug: types, the publisher's descriptions, ranges and the values each field holds. Find slugs with search_datasets or resources/list. The MCP server records nothing about who calls it or what they ask. To keep the rate limit, it counts row tool calls per address in Cloudflare's cache for 10 seconds.