read_pages

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

No annotations are provided, so the description carries the full behavioral burden. It does disclose the key trait that results are cached and that output is metadata (source, line count, mean confidence) rather than text, which is genuinely useful. But it says nothing about whether OCR is invoked, permission/auth needs, or what happens on a missing/invalid page range, so large gaps remain for a 5-param read tool.

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

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

It is a single short sentence with no filler, and the differentiating fact is front-loaded. However, the sentence is grammatically clipped ('Several pages read and cached'), which slightly impairs parsing rather than maximizing clarity.

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

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

For a tool with 5 parameters, no annotations, and no output schema, the agent still lacks guidance on dpi/lang/force_ocr behavior and on what 'cached' implies for the return payload format. The one sentence covers purpose and one differentiating detail but leaves the call-shaping decisions unresolved.

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

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

Schema coverage is only 20% – just the pages range syntax ('3', '1-5', 'all') is documented. dpi, lang, force_ocr and path have no schema descriptions, and the tool description adds no meaning for them (e.g., that force_ocr bypasses embedded text, or that lang affects OCR). With low coverage, the description is expected to compensate and does not.

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

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

The verb+resource (read pages, plural/cached) is identifiable and the parenthetical explicitly distinguishes it from the sibling read_page by saying which one returns text. The phrasing 'Several pages read and cached' is a fragment, but the intent is recoverable. No confusion with pdf_info, find, or words_in.

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

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

The contrast with read_page implies 'use this when you need multiple pages worth of metadata rather than text', but the selection rule is only implied. No explicit when-not conditions, no mention of cost/perf tradeoff versus looping read_page, and no prerequisites.

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