Skip to main content
Glama

get_page_image

Render a datasheet page, element, caption, or region as a PNG at the needed resolution so an LLM can read tables, figures, and fine print from the source image.

Instructions

The original as a PNG image: an element, a caption, a region or a full page. Read the values from this image. An element, a caption or a region is rendered at 300 to 600 dpi when it fits max_px: prefer them, because a full page costs more input. A full page is correct for a table that fills the page. It is rendered at the resolution that fits max_px. A caption with no table or figure in the element map gives its full page. Gives a line with the page, the region, the dpi and the size, and the image.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "document": "<document>"
    },
    "name": "get_page_image"
  }
}

Input Schema

NameRequiredDescriptionDefault
bboxNoOr a region [x0, y0, x1, y1] in pt, origin at the top left of the page (the units of get_page_map). Needs page.
pageNoThe page number, from 1. Not necessary with element.
max_pxNoThe maximum long edge of the image in pixels (256 to 8000), for a full page and also for an element, a caption or a region: a large element gets less than 300 dpi when it does not fit (the result gives the dpi). 1600 is enough for a table that fills the page. Use more for small print, a fine drawing, or a large table at 300 dpi.
captionNoOr a caption label, for example "Table 57" or "Figure 3-1": the image shows this table or figure. A table that continues on the next pages gives one image for each page.
elementNoAn element ID from get_page_map or list_items, for example p7.T1: the image shows this element.
documentYesThe document: its reference from list_documents, project_status or add_document (the first 12 characters of its SHA-256). A longer SHA-256 prefix or the numeric document ID also works.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changedv0.5.0
    • changedInput schema / properties / max_px / description
      Previous value: -"The maximum long edge of a full-page image in pixels (256 to 8000). 1600 is enough for a table that fills the page. Use more for small print or a fine drawing."New value: +"The maximum long edge of the image in pixels (256 to 8000), for a full page and also for an element, a caption or a region: a large element gets less than 300 dpi when it does not fit (the result gives the dpi). 1600 is enough for a table that fills the page. Use more for small print, a fine drawing, or a large table at 300 dpi."
  2. First observedv0.1.0

TDQS

A3.6/5.0
Behavior3/5

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

With no annotations, the description does real work: it discloses the 300-600 dpi rendering of elements, that dpi drops when the target exceeds max_px (and that the result reports the actual dpi), and that full pages cost more input. It stops short of failure modes (bad element ID, page out of range) or any permission/availability caveats. Solid but not full coverage of the behavioral burden.

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

Conciseness3/5

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

Every claim is on-topic, but the prose is telegraphic and article-dropping ('Gives a line with the page, the region, the dpi and the size'), and the ordering jumps from purpose to preference to exception to output. It reads like bullet fragments flattened into a paragraph, which costs readability for no length savings.

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

Completeness4/5

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

For a 6-parameter tool with no output schema, the description covers the essentials: what is returned (the image plus a line with page, region, dpi and size) and how resolution is chosen. Only error behavior and the relationship to text-extraction siblings are left out.

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

Parameters3/5

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

Schema coverage is 100%, so baseline is 3, and much of the prose (max_px fitting behavior, element ID format, caption examples) closely mirrors the schema descriptions rather than adding new meaning. The genuinely additive part is the selection guidance between element/caption/bbox/page, plus the practical hint that 1600 px suffices for a page-filling table.

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

Purpose4/5

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

The opening line names the concrete artifact (a PNG image of the original) and enumerates the four targets: element, caption, region, or full page. It also ties element IDs back to get_page_map/list_items, which helps place it in the sibling set. It never states the verb-vs-alternative relationship to get_element_text or get_pdf_pages, so an agent must infer that this is the visual fallback.

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

Usage Guidelines4/5

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

Explicit preference is given: 'prefer' element/caption/region because 'a full page costs more input,' with a stated exception (a table that fills the page), plus the caption-with-no-mapped-table case degrading to a full page. What's missing is when not to reach for this tool at all versus text-extraction siblings such as get_element_text.

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