Skip to main content
Glama
puf3zin

art-mcp

by puf3zin

Search artworks

search_artworks

Search museum collections worldwide for artworks by artist, title, culture, or period. Returns compact results with source and id for further use.

Instructions

Search museum collections for artworks. Returns compact results including a source and id used with get_artwork or get_artwork_image.

IMPORTANT — this matches keywords against catalog metadata; it is not semantic search. Artist names, titles, cultures, and periods work well. Descriptions of subject matter work badly: a query like 'man crossing a bridge' matches any record containing 'man', 'bridge', or names like 'Bridges' and 'Cross', and buries the relevant results.

So translate the user's intent into catalog vocabulary before calling. If you can name a likely artist, school, or title from your own knowledge, search for that instead — e.g. prefer 'Hokusai' or 'Hiroshige Tokaido' over 'japanese painting of a bridge'. Searching a subject phrase directly is a last resort; if you must, use one or two distinctive nouns rather than a sentence.

Results are ordered by each museum's own relevance ranking, which is often weak — scan the whole list rather than assuming the first hit is best.

CHOOSING A SOURCE. 'all' is a fine default, but naming a source is much better when the work has an obvious home. Five of the ten sources hold mostly Western art (met, artic, cleveland, harvard, rijksmuseum); for anything else, reach for these first:

  • Japanese art and objects → 'japansearch' (Japan's national aggregator), then 'vam', 'met'

  • Chinese, Korean, South and Southeast Asian, Islamic → 'vam', then 'smithsonian'

  • African art → 'smithsonian' with unit_code:NMAfA in the query, then 'vam'

  • Indigenous Americas and pre-Columbian → 'smithsonian' with unit_code:NMAI

  • Māori and Pacific → 'tepapa'

  • Latin American art, and anywhere else with no museum API → 'wikidata'

Source-specific quirks worth knowing:

  • 'smithsonian' accepts field filters in the query, so unit_code:NMAfA or unit_code:FSG (Asian art) narrows 21 museums to one. Write it as a plain AND term — 'Nasca AND unit_code:NMAI' — and do not parenthesise it, which makes the filter silently drop. Note that unit_code:NMAI returns metadata but essentially never images, so use it to find and read about Indigenous American works, not to display them.

  • 'japansearch' catalogs in Japanese and has no English titles — query it in Japanese ('北斎', '浮世絵') for good recall, and expect titles to come back in Japanese with a language field marking them; that is correct output, not a failure.

  • 'tepapa' images are mostly 'All Rights Reserved' under Te Papa's cultural protocols for taonga Māori. They can still be fetched and viewed, but check license from get_artwork before suggesting a work can be reused.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
limitNoMax results per source (default 10).
queryYesCatalog keywords — ideally an artist name, title, culture, or period, e.g. 'Katsushika Hokusai' or 'Edo period landscape'. Avoid full sentences and subject descriptions.
sourceNoA specific museum source, or 'all' (default) to search every available source. An all-source search costs several seconds — 'rijksmuseum' has no free-text search, and 'wikidata' runs a SPARQL query — so name a source when you know which one holds the work.
Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations are provided, so the description carries the full burden of behavioral disclosure. It explicitly states the search is not semantic, warns about weak relevance ranking, details source-specific quirks (e.g., smithsonian field filters dropping silently, japansearch returning Japanese-only titles, tepapa image licensing restrictions), and explains the time cost of an 'all' search.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is long but well-organized with sections (IMPORTANT, CHOOSING A SOURCE, Source-specific quirks). Every sentence adds necessary context; however, the length could slightly tax LLM context windows. It is front-loaded with the core purpose and key warnings, earning a high score but not perfect conciseness.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given the tool's complexity (multiple sources, non-semantic search, weak ranking, field filters, language differences), the description is remarkably thorough. It covers return format, behavioral quirks, and actionable guidelines for all major scenarios. The lack of output schema is compensated by explaining what results contain ('source' and 'id').

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, providing baseline meaning for each parameter. The description adds significant value by elaborating on the 'query' parameter (preferred format, translation advice), the 'source' parameter (detailed guidance on choosing sources with cultural/regional recommendations and quirks), and the 'limit' parameter (default behavior and implications).

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description begins with a clear verb-resource pair ('Search museum collections for artworks') and immediately distinguishes the tool's output from siblings by noting it returns 'source and id used with get_artwork or get_artwork_image'. It provides specific enough detail to differentiate from list_sources, get_artwork, and get_artwork_image.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description gives extensive when-to-use guidance: it explains that keyword search is not semantic, instructs translating user intent into catalog vocabulary, recommends preferring known artist/title over subject descriptions, and provides a last-resort strategy. It also advises scanning the entire result list rather than assuming the first hit.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Install Server

Other Tools

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/puf3zin/art-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server