rijksmuseum-mcp+

Overview Schema Source Related Servers Discussions Admin

Inspect Artwork Image

inspect_artwork_image

Read-onlyIdempotent

Inspect an artwork image to analyze visual details, zoom into regions, or read inscriptions. Returns image bytes the model can see and reason about immediately.

Instructions

Returns image bytes (base64) for the LLM's own visual analysis. Covers a whole artwork or a region. The LLM can see and reason about the image immediately — not for the user to view (use get_artwork_image for the interactive viewer). Not for listing or summarising artworks — use search_artwork.

Use with region 'full' (default) to inspect the complete artwork, or specify a region to zoom into details, read inscriptions, or examine specific areas. The response includes cropPixelWidth/cropPixelHeight: the actual pixel dimensions of the returned image.

Region coordinates: 'pct:x,y,w,h' (percentage of full image, recommended), 'crop_pixels:x,y,w,h' (pixel coordinates of the full image — use with nativeWidth/nativeHeight from a prior response), or 'x,y,w,h' (legacy IIIF pixels, equivalent to crop_pixels). A tight region returns more real detail than a larger size — the server never upscales past the region's own pixels. For multi-panel works, estimate panel percentages from the physical dimensions in get_artwork_details.

When a viewer is open for this artwork, it automatically zooms to the inspected region (navigateViewer defaults to true, no effect when region is 'full'), so no separate navigate_viewer call is needed for basic zoom.

The response includes the active viewUUID (if any) for follow-up navigate_viewer calls.

Input Schema

TableJSON Schema

Name	Required	Description	Default
`size`	No	Width of returned image in pixels (200–1988, default 1568). Vision models bill images in 28×28 patches, and that ceiling is the largest width clearing both the per-image patch budget and the stricter per-image limit that applies once a conversation has accumulated many images. Tall and square regions clamp below that automatically — the response reports the width actually delivered.
`region`	No	IIIF region: 'full', 'square', 'pct:x,y,w,h' (percentage), 'crop_pixels:x,y,w,h' (pixels of the full image — use with nativeWidth/nativeHeight from a prior response), or 'x,y,w,h' (legacy IIIF pixels, equivalent to crop_pixels). E.g. 'pct:0,60,40,40' for bottom-left 40%.	full
`quality`	No	Image quality — 'gray' can help read inscriptions or signatures	default
`rotation`	No	Clockwise rotation in degrees
`viewUUID`	No	Target a specific viewer session (from get_artwork_image). When omitted, auto-discovers a viewer for this artwork.
`objectNumber`	Yes	The object number of the artwork (e.g. 'SK-C-5')
`navigateViewer`	No	Auto-navigate the open viewer to the inspected region (default: true). Only effective when a viewer is open for this artwork.

Output Schema

TableJSON Schema

Name	Required	Description
`error`	No
`title`	No	Artwork title — mirrors the caption in the text/image channel so structuredContent readers retain it.
`region`	Yes
`creator`	No	Artwork creator — mirrors the caption in the text/image channel.
`quality`	Yes
`rotation`	Yes
`viewUUID`	No	Active viewer session ID (if a viewer is open for this artwork)
`warnings`	No
`cropRegion`	No	Normalized IIIF region used for the fetch; crop_pixels: inputs are normalized to plain IIIF pixel regions.
`fetchTimeMs`	No	Time spent fetching from IIIF server (ms)
`nativeWidth`	No
`nativeHeight`	No
`objectNumber`	Yes
`requestedSize`	Yes
`cropPixelWidth`	No	Actual width in pixels of the returned inspect image/crop. Use with cropPixelHeight for crop-local pixel overlays.
`regionRecovery`	No	Out-of-bounds recovery hint. Present only on an `overlay_region_out_of_bounds` error — mirrors the recovery payload that the text channel renders, so a structuredContent reader can self-correct without parsing prose.
`cropPixelHeight`	No	Actual height in pixels of the returned inspect image/crop. Use with cropPixelWidth for crop-local pixel overlays.
`viewerNavigated`	No	Whether the viewer was auto-navigated to the inspected region

Schema Changelog

Changes observed during successful MCP inspections.

Changed5 schema fields changedv0.90.1Sep 6, 2026

changedInput schema / properties / size / description

Previous value: -"Width of returned image in pixels (200–2016, default 1568). Defaults align to multiples of 28 for clean LLM coordinate handling: 1568 is Sonnet 4.6's native resolution cap, 2016 is the highest ×28 multiple that stays within Opus 4.7's per-image token budget across common aspect ratios."New value: +"Width of returned image in pixels (200–1988, default 1568). Vision models bill images in 28×28 patches, and that ceiling is the largest width clearing both the per-image patch budget and the stricter per-image limit that applies once a conversation has accumulated many images. Tall and square regions clamp below that automatically — the response reports the width actually delivered."

changedInput schema / properties / size / maximum
```
Previous value: -2016New value: +1988
```

removedOutput schema / properties / creator / anyOf

Removed value: -[
-  {
-    "type": "string"
-  },
-  {
-    "type": "null"
-  }
-]

addedOutput schema / properties / creator / type
```
Added value: +[
+  "string",
+  "null"
+]
```

addedOutput schema / properties / warnings

Added value: +{
+  "items": {
+    "type": "string"
+  },
+  "type": "array"
+}

First observedv0.90.0Jul 4, 2026

TDQS

A5/5.0

Behavior5/5

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

Annotations already mark the operation read-only, idempotent, and non-destructive, and the description adds concrete behavior: the server never upscales beyond the region's own pixels, tight regions reveal more real detail, viewer auto-navigation occurs, and the response includes crop dimensions and viewUUID. This is substantial behavioral context beyond what annotations provide.

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

Conciseness5/5

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

Front-loaded with purpose and exclusions, then organized into region behavior and viewer behavior. Despite its length, each sentence adds operational value and there is no filler.

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?

For a parameter-rich image inspection tool, the description covers target audience, region syntax, resolution behavior, viewer coordination, and relevant response fields. Since an output schema exists, return-value details do not need to be repeated here.

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

Parameters5/5

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

The input schema already has 100% parameter coverage, but the description adds practical meaning: region syntax variants are explained with usage guidance, crop_pixels should be paired with nativeWidth/nativeHeight from a prior response, tight regions give more real detail, and navigateViewer behavior is clarified. This goes well beyond the schema's descriptions.

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 opens with a specific verb and resource: 'Returns image bytes (base64) for the LLM's own visual analysis.' It immediately distinguishes the tool from get_artwork_image and search_artwork, so an agent knows exactly what it is for and what it is not for.

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?

Explicitly names alternatives and conditions: 'not for the user to view (use get_artwork_image)', 'Not for listing or summarising artworks — use search_artwork', and notes that navigateViewer removes the need for a separate navigate_viewer call. This is clear when-to-use versus when-not-to-use guidance.

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

Deploy Server