Tools

Every tool is read-only: it searches or reads Nylon’s library and changes nothing. Your assistant picks the tool; you don’t call them directly.

search_librarySearches Nylon’s library of New Zealand, Australian and UK legislation, cases and commentary.

read_documentReads within one document from Nylon’s library: the whole document, the passages that best match a phrase, or the passages either side of a known extract.

Instructions sent to the assistant

When citing a source, use ONLY the `link` field from the result. Never construct, guess, or recall a source URL yourself — in particular do not link to austlii.edu.au, nzlii.org, or other external case-law databases. If a result has no link, cite it by title without a URL.

How they work together

A typical answer takes two steps. The assistant calls search_library with one or more searches, filtered to the jurisdiction you asked about. Then it calls read_document on the most relevant results, to check the exact wording or read further. Every extract carries a link to the source in Nylon, and the assistant cites those links in its answer.

When an assistant connects, the server also sends it instructions: cite only the link field from results, and never construct or recall a source URL.

Both tools carry MCP annotations: readOnlyHint: true, destructiveHint: false and openWorldHint: false. Results come back as text, not structured content.

About the schemas

The parameter tables and schemas on each tool page are generated from the server’s own tool definitions. One difference: in production, the jurisdictions, contentTypes and practiceAreas filters are restricted to Nylon’s live catalogue, and the assistant receives the valid values in the tool definition. The search_library page lists them.